Butler University Butler University

Digital Commons @ Butler University Digital Commons @ Butler University

Scholarship and Professional Work - Business Lacy School of Business

1-1-1998

The Effect of Lifting the Blindfold From Civil Juries Charged with The Effect of Lifting the Blindfold From Civil Juries Charged with

Apportioning Damages in Modified Comparative Fault Cases: An Apportioning Damages in Modified Comparative Fault Cases: An

Empirical Study of the Alternatives Empirical Study of the Alternatives

Jordan H. Leibman Indiana University

Robert B. Bennett Butler University, rbennett@butler.edu

Richard Fetter Butler University, rfetter@butler.edu

Follow this and additional works at: https://digitalcommons.butler.edu/cob_papers

Part of the Civil Law Commons, and the Civil Procedure Commons

Recommended Citation Recommended Citation Leibman, Jordan H.; Bennett, Robert B.; and Fetter, Richard, "The Effect of Lifting the Blindfold From Civil Juries Charged with Apportioning Damages in Modified Comparative Fault Cases: An Empirical Study of the Alternatives" (1998). Scholarship and Professional Work - Business. 155. https://digitalcommons.butler.edu/cob_papers/155

This Article is brought to you for free and open access by the Lacy School of Business at Digital Commons @ Butler University. It has been accepted for inclusion in Scholarship and Professional Work - Business by an authorized administrator of Digital Commons @ Butler University. For more information, please contact digitalscholarship@butler.edu.

THE EFFECT OF LIFTING THE BLINDFOLD FROM CIVIL JURIES CHARGED WITH APPORTIONING DAMAGES IN MODIFIED COMPARATIVE FAULT CASES:

AN EMPIRICAL STUDY OF THE ALTERNATIVES

Leibman, Jordan H.

Bennett, Robert B.

Fetter, Richard

INTRODUCTION

Among the most contentious issues in American procedural law is the question of how much civil juries should be told about the legal consequences of their findings. One view holds that the jury's role should be no more than that of fact-finding tribunal having no legitimate interest in the ultimate outcome of the lawsuit.(n1) In this view, it is within the judge's function to fashion the ultimate-outcome judgment by applying the jurisdiction's law to the facts as determined by the jury. Information beyond what is necessary to try the facts in dispute should be withheld from the jury lest it be tempted by sympathy or bias to veer from an objective appraisal of the evidence to weigh it instead in accordance with a predetermined and desired outcome. In jurisdictions where this view predominates, "blindfold" rules prohibit disclosure to the jury, either by instruction or implication, of some or all of the legal consequences of the jury's findings.(n2)

Another view sees the jury as an institution of last resort with the power to exercise the conscience of the community, even in derogation of the law, if that is necessary to achieve just results in individual cases.(n3) One proponent of this view observes that many questions deemed factual are really "mixed questions that jump the artificial law/fact boundary. This is true in negligence cases, where juries decide the fact of whether a defendant's behavior fell below the behavior expected of a reasonable person."(n4) To answer "factual" questions such as this in ignorance of the answers' consequences can produce arbitrary, inequitable, and unintended results. Moreover, efforts to blindfold juries are often ineffective or counterproductive as jurors will seek to infer, correctly or incorrectly, the economic effects of their verdicts on the parties. For these reasons, opponents of blindfolding would admit the sunshine of legal knowledge into the jury room. In jurisdictions where this view predominates, we find what we will refer to as "sunshine" rules requiring or permitting the legal consequences of jury findings to be revealed to the jury.(n5)

One of the most troublesome jury disclosure problems is encountered under systems of modified comparative fault. Under modified comparative fault, plaintiffs are barred from all recovery if the trier of fact finds their percentage of fault to be equal to or above some specific threshold, generally either fifty or fifty-one percent. In this article, we refer to this threshold as the percentage bar to the plaintiffs recovery or percentage bar. In jurisdictions where there is a fifty-one percent bar to recovery, a plaintiff fifty percent at fault can still recover fifty percent of her damages; in jurisdictions with a fifty percent bar the fifty-percent-at-fault plaintiff recovers nothing.(n6) The question of whether to inform the jury about the existence of a percentage bar

is critical because a single percentage point difference in determining the degree of plaintiff's fault can have a drastic result: a half recovery versus no recovery. For that reason, blindfolding proponents argue that the percentage bar rule must be kept from the jury to reduce the effects of sympathy and bias. Opponents argue that the comparative fault jury's task of quantifying culpability requires the jurors to make moral judgments; to do so, triers of fact need to know what consequences attach to their fault determinations.

Until recently, these opposing arguments have been primarily doctrinal, focusing on procedural rule regimes likely to produce the most balanced, therefore the fairest, forum for trial.(n7) In 1995, however, as part of a national tort reform movement, the Illinois General Assembly enacted a far-reaching set of amendments to its tort statutes(n8) that were expressly designed to reduce the liability exposure of a number of constituencies often found to be defendants in tort cases.(n9) Reducing the liability burden on manufacturers and presumably other employers as well "helps them to do what they're supposed to be doing: creating jobs and helping our economy grow free from the worry of unwarranted lawsuits."(n10)

Illinois is among the many states that maintain systems of modified comparative fault.(n11) The Illinois legislature saw the disclosure rules governing jury instructions in tort cases as primarily an economic or "tort reform" issue. If jury sympathy and bias generally favor plaintiffs, a measure that reduces the potential for exercising these impulses would appear to favor defendants. Therefore, if juries know that attributing fifty or fifty-one percent of the fault for injuries to plaintiffs will bar them from all recovery, it would seem that the aggregate economic burden to defendants can be reduced by eliminating this information from the juries' deliberations. Apparently persuaded by this reasoning, the Illinois legislature, in the 1995 amendments, moved from a mandated sunshine rule to a mandated blindfold rule with respect to the existence of a percentage bar to recovery.(n12)

Legislation designed to produce an economic effect should, if at all possible, be tested empirically, preferably before enactment. To that end, we have developed and performed an experiment to compare the effect of a blindfold rule that prohibits disclosure of a percentage bar to recovery with that of a sunshine rule that requires disclosure of that information to the jury.

In this article, we begin, by way of background, with a discussion of the origins and development of comparative fault in the United States.(n13) We then trace the interesting history of blindfold rules, focusing specifically on the prohibition against disclosing the existence of a percentage bar under modified comparative fault.(n14) We then describe the experiment and conclude from its data that a sunshine rule does cause juries, on average, to increase the percentage of fault attributable to defendants, thus generating more frequent awards to plaintiffs.(n15) However, the data also indicate that sunshine juries make smaller awards, so that the aggregate economic impact of implementing the sunshine rule appears to be statistically insignificant.(n16)

The experiment also provides a comparative measure of the economic effects of the four principal defensive rule regimes: 1) contributory negligence as a complete defense;(n17) 2) pure comparative fault;(n18) 3) modified comparative fault under the blindfold rule;(n19) and 4) modified comparative fault under a sunshine rule(n20)

HISTORICAL BACKGROUND

The Origins and Development of Comparative Fault

In this article, the authors use the term comparative fault to refer to any system in which the responsibility for an injury is assigned to the parties by the trier of fact in proportion to the parties' wrongful conduct that proximately caused the injury. The species of conduct deemed to be "fault" varies from jurisdiction to jurisdiction. Fault might include acts or omissions of ordinary negligence, negligence per se, gross negligence, willful or wanton conduct, recklessness, conduct deemed tortious under strict liability principles, breach of implied or express warranty, and, in one state, intentional conduct.(n21) We use the term comparative negligence to refer to systems of comparative fault under which only conduct that can be characterized as negligence is to be compared.(n22)

Contributory negligence emerged as a complete defense in England in 1809(n23) and in the United States in 1824.(n24) The principles of damages apportionment, however, were applied much earlier. Legal historians trace the roots of comparative negligence to the Digest of Justinian, a work that sought to summarize Roman law up to 533 A.D.,(n25) but one scholar Credits the medieval sea codes for the introduction of comparative negligence into the codes of civil law jurisdictions(n26) where it became well established long before its acceptance in this country.(n27) In England, admiralty law provided for equal division of damages until 1911, and that was the rule in the United States until 1975 when the Supreme Court adopted the concept of apportionment based on fault.(n28) By 1953, the United States was virtually the last stronghold of contributory negligence.(n29)

Prosser and Keeton find a host of explanations in the authorities for the persistence of contributory negligence as a complete defense in the United States.(n30) Given that any degree of contribution by the plaintiff to his own injury barred him from any recovery, juries found it difficult to find for plaintiffs under this rule. In common accident situations, such as railroad crossing collisions, where some failure on the part of the injured party to exercise due care could be presumed, courts frequently took cases against railroads and other industries away from increasingly plaintiff-oriented juries to decide them for the defendants as matters of law.(n31) Ultimately, courts began to recognize the one-sidedhess of this regime, and limitations on the defense arose, such as the "last clear chance" doctrine, that softened its harshness in certain situations.(n32)

At least two states experimented with comparative principles in the nineteenth century. Illinois adopted judicially for a period of time the concept that a slightly at fault plaintiff could recover from a grossly at fault defendant, but no provision was made for apportionment of damages.(n33) In Georgia, an 1860 statute provided for diminution of a negligent plaintiffs damages in railroad cases.(n34) In 1913, the Georgia Supreme Court, in Elk Cotton Mills v. Grant,(n35) combined this railroad statute with another one of general applicability, which provided that "the defendant is not relieved, although the plaintiff may in some way have contributed to the injury sustained," thus creating through statutory construction a comparative negligence system for the state.(n36) The court relied on a 1904 case,(n37) which held that, in

order to recover, the plaintiffs negligence must be less than the defendant's.(n38) Thus, Georgia's was the first modified comparative negligence system in the United States.

Five years earlier, Congress had enacted the second Federal Employers' Liability Act(n39) (FELA), a "pure" form of comparative negligence to cover interstate railroad employees.(n40) Coverage for intrastate railroad workers was later enacted by thirty states.(n41) The first comparative negligence statute of general applicability was enacted by Mississippi in 1910.(n42) It also provided for the pure form, emulating the FELA.(n43) In 1931, Wisconsin enacted a general comparative negligence statute based on the FELA and its own intrastate railroad workers statute.(n44) This latter statute contained a fifty percent recovery bar, which became part of the 1931 law.(n45) Wisconsin is generally credited with developing and refining the modified form of comparative negligence.(n46)

Nebraska in 1913, and South Dakota in 1941, enacted slight/gross systems, but this approach failed to attract a following.(n47) Tennessee, in 1907, judicially adopted a provision permitting a plaintiff whose contributory negligence was "remote" to recover, but that concept has since been supplanted there by a modified, fifty-percent-recovery-bar system.(n48) Until the late 1960s, only one other state, Arkansas, enacted a comparative negligence law,(n49) despite numerous attempts to pass such legislation.(n50)

In the 1960s the fault system in the United States came under attack in two major areas: state courts began to apply strict liability principles in defective product cases,(n51) and a number of influential commentators began to promote the concept of no-fault automobile insurance laws.(n52) Professor Schwartz attributes the ensuing rush of state legislatures to enact comparative fault legislation from 1965 through 1973 to "a desire to save the fault system."(n53) Business and liability insurance interests certainly preferred the hitherto unpalatable apportionment of fault concept to the more radical no-fault proposals.(n54)

The first state court of last resort to adopt judicially a general comparative negligence system was Florida's in 1973. In Hoffman v. Jones,(n55) the Florida Supreme Court chose the pure form as did the majority of courts that followed Florida's example of judicial adoption.(n56) Today, twelve states maintain the pure form of comparative fault;(n57) thirty-three have modified systems;(n58) one state, South Dakota, still retains the slight/gross system;(n59) and four states, all in the southeast, retain contributory negligence as a complete defense.(n60)

Because comparative fault represented a compromise between the defensive regimes of contributory negligence and no-fault, it is not surprising that many state legislatures opted for the modified form. As we will demonstrate, modified comparative fault imposes a significantly smaller burden on defendants than does the pure form.(n61) Although the economic argument for modification is realistic, this system, like that of contributory negligence, permits defendants whose wrongful conduct proximately caused injury to escape liability entirely.(n62)

At the same time, however, a measure of flexibility is provided by leaving it to the jury to determine whether the percentage bar for the plaintiffs recovery has been reached. Although making this determination is an exercise in physical fact-finding, it is more than that the task requires a moral weighing of relative culpability.(n63) Additional flexibility is provided by

permitting the jury to determine the total amount of damages, an amount it can reduce if it believes it has stretched to find the plaintiffs percentage of fault below the jurisdiction's percentage bar.(n64) Although the modified comparative fault jury may have these fine-tuning tools at its disposal, the question arises whether the jury should be made aware of how it might exercise this power.(n65) The issue of whether to disclose to the jury the existence of a percentage bar can be effectively examined by considering how four state jurisdictions have resolved the disclosure question.

BLINDFOLD V. SUNSHINE IN FOUR JURISDICTIONS

The Origin of Blindfolding -- Wisconsin

The rule against disclosing to civil juries the existence of a percentage bar to recovery emerged in Wisconsin as a corollary to that state's strong preference for special verdict practice in civil cases.(n66) The purpose of special verdicts is to divide more clearly the functions of judge and jury.(n67) The special verdict charge requires the jury to submit in writing answers to a series of factual questions.(n68) What is not asked of the special verdict jury is to determine the ultimate outcome of the litigation. That task is assigned to the court, which is to fashion an ultimate-outcome judgment consistent with the special verdict answers and the applicable state law.(n69)

To secure bias-free answers to special verdict questions, the Wisconsin Supreme Court, in 1890, ruled that the jury "must not consider the effect which their answers may have... upon the controversy or the parties."(n70) This was long before comparative negligence was enacted by any state.(n71) In 1907, the court ruled that when special verdicts are used, it would constitute "reversible error for the trial court by instruction.., to inform the jury expressly or by necessary implication of the effect of [a special verdict answer] upon the ultimate right of either party litigant to recover or upon the ultimate liability of either party litigant."(n72) Thus, when the Wisconsin legislature enacted comparative negligence in 1931,(n73) the blindfold rule was already in place for cases in which special verdicts were used. This was almost always because special verdicts were--and still are--deemed "to be the rule and not the exception" in Wisconsin.(n74) In 1940, the supreme court tightened the blindfold further by holding that "counsel was also precluded from arguing the effect of its answers to the jury."(n75)

The Wisconsin model of modified comparative fault with special verdicts, fifty percent recovery bar, and blindfold was followed by the nine states adopting comparative negligence up to the 1970s.(n76) Blindfolding was the clear majority rule at that time.(n77)

The blindfold rule proved to be troublesome in practice. Juries were not easily dissuaded from seeking to produce desired ultimate outcomes with the answers to their special verdict questions.(n78) As judges strove mightily to keep them in the dark with respect to legal consequences, jurors persisted in attempting to infer what those consequences were likely to be. Prior to enactment of modified comparative fault they often inferred correctly, based on their prior knowledge, common sense, and hints derived from how the evidence was being presented and argued by counsel. To avoid reversal in these cases, courts in Wisconsin and other states were compelled to promulgate "common knowledge" exception rules.(n79) Similarly, a "conditional question" exception became necessary to deal with series of special verdict

questions that had the effect of tipping off juries as to what effect their yes or no answers might have on the ultimate outcome.(n80)

More troublesome were the cases presented to juries after modified comparative fault was enacted. Now it became less likely that jurors would infer correctly the existence and operation of the percentage bar.(n81) In states with a fifty percent bar, uninformed juries frequently elected to split the responsibility evenly between plaintiff and defendant, an attractive proposition for juries that found it difficult to assign percentages of fault with scientific precision.(n82) Because this scenario operated systematically against plaintiffs, Wisconsin raised its percentage bar from fifty to fifty-one percent, but it retained the blindfold rule.(n83)

Other states dealt with these problems by giving trial judges considerable discretion whether or not to blindfold in individual cases.(n84) Others concluded that the blindfold concept represented a failed experiment and had to go.(n85)

The Blindfold Rule is Rejected -- Colorado

Colorado was one of the nine early adopters of the Wisconsin model of modified comparative fault.(n86) But in 1974, in a Colorado appellate decision, the court rejected the Wisconsin approach, calling instead for a mandated sunshine rule. In Simpson v. Anderson,(n87) the court applied a rule that recognized that "jurors collectively represent a cross-section of the conscience of the community."(n88) Jurors should be trusted and courts must recognize that, realistically, "juries will anticipate the consequences of their findings relative to the percentage of negligence."(n89) If so, it is "preferable for the jury to deliberate with an understanding of the true effect of the law rather than under misapprehensions."(n90) Recognizing that the manner in which the law applies "will ultimately become known to at least some members of the community who will be asked to sit upon juries,"(n91) it is "far better for courts to be the vehicle by which the law is explained."(n92) Simpson was the first appellate decision in the country to reject the blindfold rule on the percentage of negligence issue.(n93)

The Colorado Supreme Court reversed Simpson.(n94) In a companion case the supreme court opined that the Colorado legislature had "intended to establish a system in negligence cases which divides the responsibility between judge and jury. Such a system enhances the chance of a pure verdict on material facts alone."(n95) Jury members need understand no law beyond that "which enables them to answer the specific questions asked of them on special verdict forms. Under this system, it is unnecessary for the jury to concern itself with how much the plaintiff receives."(n96)

The two Colorado opinions succinctly set forth the doctrinal positions of the sunshine versus blindfold debate. The Colorado Court of Appeals argued that, whether or not juries are informed of legal consequences, it is in their nature to attempt to influence the ultimate outcome of the cases upon which they sit. Their findings are less likely to be distorted by mistaken legal conclusions if they understand where their findings may lead. The Colorado Supreme Court's position was that juries can be effectively limited to narrow fact-finding without regard to ultimate outcomes, and should be required to do so in accord with the legislative intent of the comparative fault statute. For Colorado the issue was resolved shortly thereafter: the legislature

sided with the court of appeals by mandating a broad sunshine rule as to the existence of a percentage bar in comparative fault cases.(n97)

There were no new adopters of the blindfold rule until 1995.(n98) Apparently, policy makers in the large majority of modified comparative fault states concluded that juries either can not or should not be dissuaded from trying to help fashion the ultimate outcome of their deliberations.

The Sunshine Rule is Reinforced -- Wyoming

Wyoming's first comparative fault statute was enacted in 1973.(n99) The language of section 1-7.2 followed closely the Wisconsin law with its special verdict requirements?(n100) In Woodward v. Haney,(n101) the Wyoming Supreme Court ruled that, when the legislature adopted Wisconsin's statutory language, it intended to adopt Wisconsin's judicial construction that held it would be reversible error to advise the jury, in argument or by instruction, of the effect of its verdict.(n102) One commentator, writing two years before Woodward was decided, had argued that the blindfold rule should not be binding on Wyoming because Wyoming's special verdict procedure differed from Wisconsin's in that Wyoming juries "can be told of the legal effect of all issues not put in special verdict form."(n103) Under Wyoming procedure only questions of damage apportionment and amount of damages were to be in special verdict form; all other issues could be argued to the jury as would be true if a general verdict were submitted. "This being the case, it is senseless to inform the jury about the effects of their unrecorded findings and at the same time conceal the applicable law on those findings which are recorded in special verdict form."(n104)

The issue was resolved when, in 1976, the Wyoming State Legislature enacted section 1-7.7, which provided that the ad damnum clause in pleadings for damages shall not be construed to prevent argument to the court or jury concerning the amount of any party's claim; and that "the court shall inform the jury of the consequences of its verdict."(n105) In Johnson v. Safeway Stores, Inc.,(n106) the Wyoming Supreme Court held that section 1-7.7 could not be harmonized with the Woodward court's plain meaning interpretation of section 1-7.1.(n107) In Johnson, section 17.7 was interpreted as a sunshine rule superseding by implication the blindfold rule of 1-7.1.(n108)

The Johnson court noted that the Colorado Legislature had amended its statute using language similar to that of section 1-7.7 and the Colorado Supreme Court had construed the provision as "imposing an independent duty upon the trial court to instruct the jury on the comparative-negligence statute's effect" even though plaintiff's counsel did not request the jury be told.(n109) The Johnson court held that the same independent duty to inform would apply to Wyoming courts.(n110)

In 1986, the comparative fault statute was amended. New section 11-109(b) retreated from the Johnson court's "independent duty" holding. The new section provided that the court was under a duty to "[i]nform the jury of the consequences of its determination of the percentage of fault" only if requested by any party.(n111) In 1994, the statute was amended once more. This time the Wyoming Legislature returned to the Johnson court's earlier holding: "The Court shall: (i) If a

jury trial ...(B) Inform the jury of the consequences of its determination of the percentage of fault."(n112)

Why this reaffirmation of the trial court's independent duty to disclose? We suggested earlier that policy makers might reason that the blindfold rule favors defendants;(n113) perhaps Wyoming legislators were seeking to benefit plaintiffs by reinforcing the state's sunshine rule. Not very likely. A commentator characterized the 1994 amendments as follows: "Whatever the intent behind the new act it represents a significant victory for defendants, and a significant loss for plaintiffs...."(n114) Although Wyoming's 1994 comparative fault law can fairly be characterized as a tort reform statute, blindfolding civil juries on the percentage of negligence question was not part of it. We suspect that the Wyoming Legislature decided that there was insufficient evidence that blindfolding would benefit the Wyoming economy sufficient to overcome the anomalous and inconsistent verdicts often rendered by uninformed jurors. To make certain that the latter result was avoided, the Legislature mandated that all modified comparative fault juries in Wyoming will deliberate in sunshine.

One Wyoming commentator discussed the potential economic impact of a sunshine rule as follows:

Although considerable attention and litigation have been devoted to the issue of informing the jury, there is no empirical evidence on whether advising a jury of the consequences of its allocation of fault is beneficial to the plaintiff or the defendant(s). For now, at least, the issue in Wyoming is moot. The new Wyoming statute requires the court to instruct the jury as to the consequences of its decision.(n115)

One jurisdiction decided not to wait for empirical evidence.

The Reemergence of the Blindfold Rule -- Illinois

Probably no state has had more experience with the various forms of comparative fault than Illinois.(n116) Over its history, Illinois has experimented with a "slight/gross" system;(n117) experienced a failed attempt by an appellate court to adopt judicially the Wisconsin model of modified comparative negligence;(n118) saw a successful but short-lived judicial adoption of pure comparative negligence;(n119) legislated a modified comparative fault system with a mandated sunshine rule;(n120) and, most recently, enacted a modified system with a mandated blindfold rule governing the percentage of negligence question.(n121)

In 1858, the Illinois Supreme Court held, in Galena & Chicago Union Railroad v. Jacobs,(n122) that the "degree of negligence must be measured and considered and wherever it shall appear that the plaintiff's negligence is comparatively slight, and that of the defendant gross, he shall not be deprived of his action."(n123) Under this slight/gross system there was no damages apportionment; either plaintiffs recovered or they did not.(n124) Later cases permitted plaintiffs more than slightly at fault to recover so long as their negligence was greatly outweighed by that of defendants.(n125) The system persisted for three decades, near the end of which, cases were decided that ignored or limited the slight/gross defense.(n126) In Lanark v. Dougherty,(n127) the

supreme court ended this experiment with comparative principles by reestablishing contributory negligence as a complete defense.(n128)

In Illinois and elsewhere, growing dissatisfaction with the harshness of contributory negligence led to common law exceptions that mitigated its effects in specific situations.(n129) Although there was considerable interest in comparative fault principles as a potentially more equitable risk distribution mechanism during the first sixty years of the twentieth century,(n130) only the Congress and a few state legislatures were able to enact statutes.(n131) Despite a number of attempts the Illinois legislature failed to produce a law.(n132) But as the fault system came under attack in the 1960s, the hitherto effective resistance of the insurance industry to comparative principles diminished.(n133) The legislative logjam in some states gave way,(n134) and in others, the courts led the way by judicial adoption of comparative fault.(n135) Their reasoning was that, since contributory negligence was a common law defense, courts had the power and duty to replace it with a more equitable rule.(n136) When the Florida Supreme Court judicially adopted comparative fault in 1973,(n137) it was followed by several other high courts including the influential California Supreme Court.(n138)

Six years prior to Florida's ground breaking adoption, an Illinois Court of Appeals, in Maki v. Frelk,(n139) attempted to apply the Wisconsin model to that case.(n140) Maki was reversed despite favorable commentary in academic circles.(n141) The supreme court rejected the notion of judicial adoption, insisting that such a change required legislation.(n142) When in the next thirteen years there still was no statute, the supreme court, in Alvis v. Ribar,(n143) followed the Florida precedent by judicially adopting the pure form of comparative fault.(n144) In response to criticisms that judicial adoption generally fails to address detailed issues, the court stated: "We believe that the use of special verdicts and special interrogatories will serve as a guide to assist the jury in its deliberations. We leave the resolution of other collateral issues to future cases."(n145) With respect to the collateral issue of whether to blindfold juries on the percentage of negligence question, the issue concerning us in this article did not arise, because under the pure system, there is no percentage bar about which the jury need be informed.

The Illinois General Assembly finally enacted a modified comparative fault statute in 1986.(n146) The section on jury instructions provided for a mandated sunshine rule: "The court shall instruct the jury in writing that the defendant shall be found not liable if the jury finds that the contributory fault of the plaintiff is more than 50% of the proximate cause of the injury or damage for which recovery is sought."(n147) As with the other three states we have examined, there is no evidence that the Illinois General Assembly ever considered economic impact when it chose its rule governing disclosure to the jury. Rather, the law makers appear to have decided this issue based on how they believed jurors would behave under both rules, choosing the one they believed would enable jurors best to discharge their obligation to find facts based solely on the evidence.

In the 1994 elections, however, American voters signaled their disenchantment with excessive governmental intrusion into the private sector. A call for legislation lifting the burden of tort litigation was presented by conservative politicians to the voters as an essential element of that agenda.(n148) In a number of states the votes for substantial "tort reform" were available.

Probably, the most far-reaching bill to be enacted was that of Illinois.(n149) In a published statement, an Illinois Senate sponsor of the legislation made clear that the Act's objectives were to reduce the costs of liability and litigation for municipalities, small businesses, farmers, not-for-profit organizations, doctors, hospitals, manufacturers, and taxpayers generally.(n150) The statement decried the "emotionalism" so often invoked on behalf of plaintiffs; readers were urged to consider that "[t]he public debate over tort reform has largely ignored the costs of being a defendant."(n151)

The 1986 section in the Civil Code on "jury instructions in tort action"(n152) was apparently one of many provisions deemed ripe for reform. The mandated sunshine rule on percentage of negligence was replaced with a mandated blindfold: "[T]he court shall not inform or instruct the jury that defendant shall be found not liable if the jury finds that the contributory fault of the plaintiff is more than 50%."(n153) This about-face by the Illinois General Assembly was the first time since the 1970s that a jurisdiction has imposed a blindfold rule with respect to the existence of a percentage bar to the plaintiff's recovery.(n154)

The new section on jury instructions was not entirely a jury blindfold. To underscore that the objective of this section was to achieve an economic result, rather than to protect the integrity of special verdicts or maintain the appropriate roles of judge and jury, the section requires the judge to "inform the jury" that any award of compensatory or punitive damages will not be taxable.(n155) One commentary predicted that the "overwhelming effect of this change will be to reduce awards."(n156) As for the percentage bar blindfold, these commentators observed that "[i]t must be surmised that the reason behind the change is that a sympathetic jury would be less willing to find a plaintiff more than 50 percent at fault and would manipulate the percentage of fault were it told the law."(n157)

To determine whether juries in the aggregate will in fact manipulate the percentage of fault (and the amount of damages), the authors have designed and performed the experiment described in the next part of this article.

COMPARING BLINDFOLDED MODIFIED COMPARATIVE FAULT VERDICTS WITH SUNSHINE VERDICTS: AN EMPIRICAL LABORATORY EXPERIMENT

Overview

The authors concluded that field research into the economic effects of blindfold and sunshine jury instructions was not feasible. Actual cases differ too much on their facts, on the quality of physical evidence, the personalities and abilities of attorneys, and the appeal and credibility of witnesses. On the other hand the question appeared researchable using laboratory methods of social psychology.(n158)

The authors scripted, produced, and videotaped two versions of a simulated, personal injury trial conducted under modified comparative fault rules.(n159) In the Baseline (Base) version the judge instructs the jury under a blindfold rule; in the Variant, the jury receives a sunshine instruction.

Groups of mock juror subjects were recruited to view each of the versions. After viewing the trial tape and receiving instructions, the groups were randomly assigned to six or seven person juries.(n160) The juries retired to separate rooms and deliberated for up to two hours, rendering verdicts on forms also calling for answers to special interrogatories.(n161) The interrogatories asked for the total amount of damages without regard to any contributory negligence; the percentage of negligence attributable to each of three parties (two defendants and the plaintiff); and the amount of damages attributable to each party to be calculated by the jury's multiplying each party's percentage of fault by the total amount of damages suffered by the plaintiff. The sum of the percentages of fault for the three parties had to equal 100%.

Following each jury's deliberations, each juror filled out an exit survey which included statements to which the individual jurors indicated levels of agreement on a five point scale ranging from strongly agree (1) to strongly disagree (5). The survey asked an additional seven demographic questions, and several open-ended questions about factors influencing the juror's personal decision and that of his jury panel.

The verdict data from the Base and Variant juries were tabulated and analyzed. Descriptive statistics (means, medians, and standard deviations) were calculated and those of the two versions were compared. Statistical tests were performed to determine whether the differences between the Base and Variant measures of central tendency were significant.

The Simulated Trial Scenario

To ensure realism, the authors felt it important to use testimony and physical evidence from an actual trial. In addition, an actual case provides a real-world benchmark verdict with which the experimental verdicts can be compared. We sought a modified comparative fault case that turned primarily on issues of fact rather than law; one in which credible evidence could be adduced at trial upon which reasonable jurors could differ as to whether the plaintiff was more or less than 51% at fault; and one in which there were multiple defendants so that the option of shifting fault among the defendants by the jury could be studied. We required a negligence case that lawyers and lay persons would likely agree was typical, and because this was a simulation, we felt it important that the personal injury be grave to ensure that the subject jurors would take the exercise seriously. After a lengthy search, we selected for our model an Indiana motor vehicle accident case which met all these requirements.(n162) We captioned the simulation: McKey v. Torino Pizza Co. Inc.

With respect to the facts of McKey v. Torino Pizza, the Indiana Court of Appeals, in discussing the underlying case upon which McKey was based, said: "Nearly every fact related to the accident is vigorously disputed." What is undisputed in the simulation is that about 7:00 A.M. on an early November day in rural southern Indiana, Dawn McKey, a sixteen year old attractive and popular cheerleader on her way to school, pulled out from a county road onto the preferred state road on which a truck called a "stepvan," driven by an employee of Torino Pizza Co., was approaching. In both the Base and Variant versions Torino Pizza is represented to the juries as "a large national manufacturer of food products." As a result of the ensuing collision, Dawn suffered massive brain injuries, which left her with no memory of the accident and virtually no subsequent ability to remember events from day to day. The only other accident witness was the

truck driver, Will Drummond, who incurred superficial injuries that were not at issue in the case.(n163)

The intersection of County Road 1040 and State Road 38 is in a valley between two low hills. Because the hills' sightlines in both directions were restricted for Dawn, her making a left turn onto S.R. 38 was fraught with danger. The State of Indiana had placed a 40 M.P.H. "advisory-speed" sign below the crest of the hill over which Will Drummond had just come, but the speed limit along S.R. 38 was 55 M.P.H. throughout the area. Advisory speed signs in Indiana are treated like other warning signs such as "yield" or "deer crossing." The statute requires drivers to take action in adherence to these warnings only when conditions indicate there is reason to do so. The parties hotly disputed the speed of the stepvan at the moment of collision, and vigorously argued whether conditions existed requiring the stepvan to slow down to 40 M.P.H. The plaintiff claimed Torino Pizza, through its driver, was negligent because he had approached the intersection at an unreasonable speed; he had failed to abide by warning signs; he was unable to control his vehicle to avoid the accident; and had failed to maintain a proper lookout.

The State of Indiana was also a defendant on the plaintiffs theory that it was well known that the intersection was highly dangerous, this notice creating a duty for the State to do more than it had done to protect drivers attempting to make left turns onto the state road The State does not appear at trial, however, because, unbeknownst to the jury, the plaintiff and the State had entered into a "covenant not to execute."(n164) Despite hearing no defense from it, the jury was permitted to allocate a percentage of fault to the State.

The trial simulation breaks down into 14 segments:

1. Plaintiffs Opening Statement: The plaintiff's attorney tells the jury about Dawn, her family, her aspirations, her injuries, suffering, and what she has lost as a result of the accident. He describes the plaintiffs theory of the case and gives reasons why the jury should allocate fault to the defendants. He treats the State's liability lightly;(n165) the dangerousness of the intersection he raises primarily to show that Torino's agent, Will Drummond, had notice of it and failed to take it into account.

2. Defendant's Opening Statement: Torino's attorney admonishes the jury not to let sympathy for Dawn affect their consideration of the liability issues. He tells jurors that Dawn should have been instructed by her parents to take a longer but less dangerous route to school that morning. He asks them to accept that Will Drummond had no reason to go on high alert at the intersection, that his driving behavior was reasonable, and that his speed approaching Dawn's car could not really be ascertained, even by Will. He asks them to attribute 100% of the fault to Dawn and the State. Finally, he promises the jury evidence that Dawn is not injured as severely as the plaintiff makes out.

3. Testimony of Carter Stumpf: The Lincoln County Sheriffs Captain is shown photographs of the road and the 40 M.P.H. advisory sign. He testifies to the well-known dangerousness of the intersection. He is shown color photographs of the wreckage with which he describes the scene at the site shortly after the accident. His accident report which reports dry pavement, daylight, and Dawn's failure to yield the right of way is put in evidence. He testifies there were no skid marks prior to collision suggesting that, because the truck driver was inattentive, he failed to brake.

4. Testimony of Will Drummond: Truck driver Drummond testifies about his experience driving grain trucks; the nature of his job as a delivery/route salesman for Torino; his suspension for tardy paper work; his return to work the day of the accident; and Dawn's pullout that day leaving him no opportunity to take evasive action. When asked about the 40 M.P.H. advisory and school bus signs, he testifies that he does not remember them being there. He claims he was not distracted by his suspension. He concedes that he was doing 55 to the crest of the hill and believes he had "eased off" the accelerator on the way down, but he can not remember his approach speed. (In an earlier deposition, he had stated he was going about 50 M.P.H. down the hill.) He testifies for the first time that Dawn stopped at the stop sign located 26 feet from the state road, then came on out without stopping again from that position. He maintains that he did brake before the collision.

5. Testimony of Vincent Schneider: Plaintiff's expert accident reconstructionist testifies that the distance the vehicles travelled after the collision on a dry pavement indicates an impact speed of 50 M.P.H. He introduces an animated computer simulation of the accident demonstrating graphically that, if Drummond had obeyed the advisory speed sign, there would have been a clean miss of 62 feet. His animation assumes Dawn stopped for the first and last time at the edge of the state road, not at the stop sign. He assigns 100% of the fault of the accident to Torino (Drummond).

6. Testimony of Neil Sundstrom: Plaintiffs expert authority on truck-driving safety testifies that Drummond's safety training was woefully deficient, but no training requirement laws were violated by Torino. He testifies that a trained driver would have immediately slowed and covered his brakes as soon as he saw the conflicting traffic (Dawn's car) moving toward the state road from the stop sign. He accepts Drummond's testimony and assumes she stopped there.

7. Testimony of Penney McKey: Mother of the plaintiff testifies as to the horrendous burdens the family has suffered because of Dawn's slow and painful physical rehabilitation and her permanent mental impairments, seizures, and anti-social personality disorders.

8. Testimony of Carla Trammell: Plaintiff's expert clinical neuropsychologist summarizes Dawn's post-accident medical history, and describes her current and permanent loss of mental function. Trammell also operates Indiana's only rehabilitation facility for these types of injury. She testifies as to the costs of a rehabilitation program for Dawn.

9. Testimony of Arnold Eagleton: Plaintiffs expert economist projects Dawn's high and low range of economic damages for lost income (based on her probable work-life expectancy) and for the costs she will incur for lifetime residential care (based on her full life expectancy). To these he adds medical and rehabilitation costs. Following Eagleton's testimony, the plaintiff rests

10. Testimony of Stanton Kramer: Defendant's expert accident reconstructionist testifies that his "momentum analysis" of the physical evidence yields an impact speed of 30-35 M.P.H. He opines that there was no time, even at that speed, for Drummond to react. He testifies that only a very cautious driver would have reacted to Dawn's moving from the stop sign. Following Kramer's testimony, the defense rests.

11. Plaintiffs First Closing Argument: Plaintiff's attorney reviews his theory of the case and how the plaintiff witnesses' testimony supports it. He admits some fault on Dawn's part, but argues that Torino and the State were more responsible. With respect to damages,

plaintiff's attorney recapitulates the figures of the economist. He adds a range of non-economic damages (pain and suffering) based on $2-$4 an hour, which Dawn will incur over her life expectancy.

12. Defendant's Closing Argument: Attorney again urges the jury not to be swayed by sympathy. He attacks the plaintiff's accident reconstruction, which assumed that Dawn stopped once at the edge of the state road. He emphasizes Drummond's good driving record and argues that Will had no reason to be rushed or distracted by his job or his suspension. He focuses the jury's attention on Dawn's failure to look left again before she turned left. He emphasizes that State Road 38 is a preferred road to County Road 1040. He misrepresents the amount of miss shown on the animated simulation as two rather than sixty-two feet. He attacks the assumptions of the plaintiff's damages testimony, suggesting that Dawn might not have been headed for college, nor is she certain to be unemployable, or need lifetime residential care.

13. Plaintiffs Final Closing Argument: In this rebuttal argument, plaintiff's attorney explains to the jury that the assumptions about Dawn's stopping point do not affect Dawn's case; either assumption leads to a "clean miss" if Drummond had just slowed down or braked. He denies defense attorney's two-foot-miss assertion. He emphasizes that, despite having the right of way, a driver on the preferred road is negligent if he fails to react to conflicting traffic moving from a dead stop in the direction of the state road. He derides Torino's failure to call experts to rebut plaintiff's damages experts, and demands that the jury accept plaintiff's uncontested evidence. He attacks the big company mentality that denies having made mistakes when they cost the company money. He concludes with an emotional description of Dawn's loss.

14. Jury Instructions: Judge Barton reads aloud 22 instructions taken from the underlying case, copies of which each juror has been given. They are to read the instructions silently along with the judge and are to bring them into the deliberation rooms. Several instructions, taken from Indiana Pattern Instructions, define "comparative fault," "negligence," "agent of the defendant," "preponderance of the evidence," "burden of proof," "proximate cause," "ordinary care," and "sole proximate cause." The judge reads the relevant Indiana statutes, and sets out the various common law duties. He instructs the jury how both are to be applied. He lists possible items of damages and informs the jury. that Dawn's life expectancy is 63.7 years. He explains the nature of expert opinion and tells the jury it must render an unanimous verdict.

The trial of the underlying case took thirteen days and generated 4,000 pages of trial transcript. The McKey simulation runs about four hours. The number of witnesses was reduced, rebuttal evidence was incorporated into direct and cross-examination, long medical depositions were summarized and integrated into one witness's testimony, and all motions, objections, and sidebars were deleted. Despite the simplification and compression, we believe that all significant factual issues were preserved.

Subjects

Mock juror subjects were 489 undergraduate students recruited on seven college campuses in Indiana, Illinois, Ohio, Oklahoma, Utah, and Washington state. The large majority of the students were sophomores or juniors either enrolled or expected to enroll in their institutions'

business programs. They had volunteered for the McKey exercise in conjunction with their introductory business law course. At this point they had either minimal or no exposure to comparative fault principles.

The pool of jurors was about equally split across genders (51% male, 49% female). The average age was 22.7 years,(n166) ranging from seventeen to fifty-seven.(n167) Racially, 81% of the jurors were white. Among the minorities were Asian (7%); African-American (4%); and Hispanic (4%).

On each campus, groups ranging from twelve to ninety-six viewed either the Base or Variant trial simulation. In most cases, the exercise was administered on evenings or weekends as an extra credit project for which the students received extra credit points. In two instances the exercise was run as a regular classroom exercise for regular credit. After viewing their versions, the students were randomly assigned to jury panels to deliberate and render comparative fault verdicts. All iterations took place during the period from July 2, 1996 to January 22, 1997.

Manipulating the Independent Variable: "Lifting the Blindfold"

The Base version of McKey is conducted as if the jurisdiction (Indiana) maintains a system of pure comparative fault. INSTRUCTION No. 1 instructs the jury to find the percentages of fault attributable to each party, to find the total damages, and to multiply the percentages for each defendant by the total damages to determine the award due the plaintiff. The verdict form conforms to this instruction. No reference is made by either attorney to the fifty-one percent bar to recovery.

In the Variant version, INSTRUCTION NO. I instructs the jury to find the percentages of fault attributable to each party, and, if more than 50% is found attributable to the plaintiff, "no further deliberations are required."(n168) This instruction conforms to Indiana law and was given in the underlying case. In the McKey simulation, however, the Variant juries also receive a SPECIAL INSTRUCTION NO. I that instructs them that, notwithstanding the instruction to cease deliberating upon finding the plaintiff 51% or more at fault, they are "specially instructed to continue [their deliberations] in each instance until [they] have agreed on a percentage of fault for all three parties." The purpose of the SPECIAL INSTRUCTION NO.1 is to obtain all the Variant percentages to compare with those of the Base juries.

In the Variant, the plaintiffs lawyer in his first closing argument states:

So, in considering all of those causes, if you find that the fault of Dawn in proximately causing this accident was more than 50%, then she loses. If you find, however, that her fault was 50% or less, and the combined fault of Torino and the state of Indiana more than 50%, then she is entitled to damages. That is the Indiana law concerning responsibility and liability.(n169)

In his final closing argument, the plaintiffs attorney states: "[Judge Barton] will instruct you to compare the duties of the parties to determine their shares of the fault involved. And if you find that Dawn McKey's fault was less than 51%, then you move to the question of damages."(n170)

INSTRUCTION NO. 1, SPECIAL INSTRUCTION NO. 1, and the closing argument references to the percentage bar are the tools with which the sole independent variable in the McKey exercise is manipulated: In the Base version, jurors are blindfolded; in the Variant, the blindfold is lifted. Dependent Variables (DV); Hypotheses (H); and Experimental Results (ER) Based on the Verdict Data

In the experimental framework that has been described, the subject jurors can react through the group behavior of their jury panels--to the manipulation of the independent variable in five ways: 1) They can choose not to reach an apportionment verdict, in which case their panels will be deemed to be "hung";(n171) 2) they can change the percentage of fault they attribute to one of the defendants; 3) they can change the percentage of fault they attribute to both defendants, which, since total fault must add up to 100% implies an inverse change in the percentage of fault attributed to the plaintiff, Dawn McKey; 4) they can change the amount of damages awarded to Dawn; 5) they can decide that no changes in percentages of fault or amount of damages are indicated.

With respect to option 5, we should point out that the application of a blindfold or sunshine rule is likely to have a significant effect only in cases where the plaintiff's fault is arguably at or near the 50% level. When the plaintiffs fault is demonstrably either slight or gross, juries will be less likely to be motivated by their knowledge of a percentage bar; their findings under either rule are likely to be similar. Policy makers who see the adoption of the blindfold rule as a tort reform issue affecting their jurisdictions' business climates should consider that many, if not most, modified comparative fault cases will be unaffected by a blindfold or sunshine rule. The McKey case, however, is one of those cases where reasonable jurors can find the plaintiff either more or less than 51% at fault.(n172) These are the cases most likely to go to a jury. on liability issues, and the ones most likely to have a precedential effect.

To study how juries generally are likely to react to this manipulation, we presented the Base (blindfold rule) version of the McKey trial to one sample of 39 jury panels and the Variant (sunshine rule) version to another sample of 45 panels.(n173) Indiana is a state in which the fault of the plaintiff is compared with the combined fault of all defendants taken as a unit.(n174) Also, Indiana is a state that permits recovery when the plaintiff and defendants are equally at fault.(n175) The McKey case is tried under these rules.

The verdict data are presented in Tables 1, 2, and 3. Table 1 presents the "raw" data from the Base juries, i.e., before the court adjusts the verdicts to conform to Indiana's fifty-one percent recovery bar. Table 2 presents the same data after the court has adjusted the Base verdicts to reflect $0.00 of damages in those instances where the jury found Dawn more than 50% at fault. Table 3 presents the data from the Variant verdicts. These require no adjustment by the court because the jury knows that a finding of 51% or more fault for Dawn means no damages for her and no liability for either defendant; under Indiana procedure the jury makes its own adjustments and renders an ultimate-outcome general verdict. Means, medians, and standard deviations for the data are indicated.

Differences in the following six dependent variables were correlated with the manipulation of the independent variable.

DV1: The mean percentage of fault attributable to both defendants combined.

DV2: The mean percentage of fault attributable to the corporate defendant, Torino Pizza Company.

DV3: The mean percentage of fault attributable to the State of Indiana.

DV4: The mean percentage of fault attributable to Dawn by those juries in the samples who did not produce an ultimate award for Dawn.

DV5: The frequency of the plaintiff's recovery.

DV6: The mean dollar recovery of the plaintiff from all juries in the samples.

The authors developed the following hypotheses predicting how the dependent variables will respond to the lifting of the jury's blindfold. We give our rationales for making these predictions, and we report the actual variations from the experimental verdict data.

H1: The mean percentage of fault attributed by the blindfold juries to both defendants combined will be less than that attributed by the sunshine juries.

Rationale: If knowledgeable juries under a sunshine rule are motivated by sympathy for a badly injured plaintiff, or by dissatisfaction with the concept of modified comparative fault, and they should wish to provide Dawn with an ultimate award, they can increase the percentage of fault attributable to the defendants so that Dawn's fault will be proportionately lowered.

ER1: The blindfold juries found the defendants to be 34.4% (mean) at fault;(n176) the sunshine juries 44.0% at fault.(n177) A t-test difference of means was applied to the data.(n178) The observed t-value was 2.26 (d.f.=80) and the observed p-value was .013. The t-test difference of means strongly supports the assertion that the Variant sunshine juries allocated considerably more fault to the defendants than did the Base blindfold juries.

H2: The mean percentage of fault attributed to Torino Pizza by the sunshine juries will be greater than that attributed by the blindfold juries.

Rationale: If sympathy and dissatisfaction with modified comparative fault are present, those attitudes should be manifested by the juries' attributing a higher percentage of fault to the corporate defendant.

ER2: The mean percentage of fault for Torino rendered by the blindfold sample was 16.3%;(n179) by the sunshine sample, 29.3%.(n180) H2 is supported by the data, based on the t-test difference of means. The observed t-value was 3.53 (d.f.=80), p-value .0003. The sunshine juries attributed nearly twice the fault to Torino than that attributed by the blindfold juries.

H3: The mean percentage of fault attributed to the State by the sunshine juries will be higher than that attributed by those in the blindfold sample.

Rationale: If sympathy and dissatisfaction are present, the sunshine juries, on average, will register a higher percentage of fault for both defendants to manifest their negative reaction to the percentage bar. To effectively lower Dawn's percentage of fault, jurors are likely to raise the percentages of fault attributable to both Torino and the State.

ER3: The blindfold juries attributed 18.1% of fault to the State,(n181) the sunshine juries attributed 14.7%.(n182) H3 is not supported by the data. The percentages of the State's fault are essentially the same for both samples, based on the t-test difference of means: t-value 1.09 (d.f. 80), p-value=.14. There is a significant probability that the observed difference, 18.1% versus 14%, is due to chance, rather than a true difference between the populations. We think there is a reasonable explanation for why the sunshine juries did not apportion more fault to the State than did the blindfold juries. Those sunshine juries who may have been motivated by sympathy and dissatisfaction would be expected to examine -- or perhaps reexamine -- the evidence in the manner suggested by the plaintiffs attorney. In McKey, the plaintiff's attorney focused almost entirely on the allegedly negligent driving of Totino's truck driver, and Torino's attack on the State was also perfunctory.(n183)

H4: The mean percentage of fault attributed to Dawn by those blindfold juries that did not produce an ultimate award for her will be greater than the mean percentage attributed to her by those sunshine juries that did not produce an ultimate award for her. Rationale: The proposition we are examining here is that the majority of sunshine juries that concluded that Dawn's negligence must bar her from recovery under modified comparative fault principles will nevertheless manifest tensions within their jury panels by registering lower percentages of fault for the plaintiff than their blindfolded counterparts.

ER4: Consider Tables 4 and 5. The blindfolded juries that produced no ultimate award for Dawn found her to be 78.5% at fault;(n184) the sunshine juries under the same condition found her to be 65.9% at fault.(n185) H4 is supported by the data, based upon the t-test difference of means (t=3.73, d.f.=80, p=.0002). This result also suggests that, while sunshine juries might wish to compensate a more-at-fault-plaintiff, their oath to find facts objectively remains a powerful constraint on the ultimate outcome of their verdicts. Further evidence that a significant number of jurors and panels were influenced by knowledge of the 51% percentage bar is found in the answers to two "manipulation check" items from the Variant sunshine exit survey. Of the 247 Variant jurors, 37.6% agreed or strongly agreed with the statement: "Because I knew that assigning more than 50% of the fault to Dawn would bar her from receiving any damages, I was influenced to allocate an increased percentage of fault to one or both defendants."(n186) Of these same jurors 37.3% reported their jury panels were so influenced.(n187)

H5: The plaintiff will recover more often under a sunshine rule than under a blindfold rule.

Rationale: Sympathy and dissatisfaction, if present, are likely to be manifested by more of the knowledgeable sunshine juries finding Dawn 50% or less at fault than blindfolded juries who are ignorant of the percentage bar.

ER5: Of the blindfold juries 30.1% found Dawn less than 51% at fault;(n188) 40% percent of the sunshine juries did so.(n189) We applied the z-test "difference of proportions"(n190) to the data.

Based upon the observed z-value (.85) and the corresponding p-value (.197), we can conclude the difference in the observed proportions (30.1% v. 40%) is sufficiently large to consider the populations of interest -- blindfold and sunshine juries -- different from one another. However, in this instance, we are about 80% confident that our conclusion is correct --i.e., that Dawn will recover more often under a sunshine rule (rather than the 95% confidence supporting H 1, H2, and H4). We should also point out that the McKey scenario poses a severe challenge for juries motivated to make an award to the plaintiff. The blindfold juries found Dawn to be 65.6% at fault.(n191) Sunshine juries would have to allocate sixteen percent more fault to defendants for Dawn to recover. We suspect that, for sunshine scenarios in which the plaintiffs' fault would be perceived under a blindfold rule to be closer to fifty percent, there would be a larger proportion of plaintiff victories. In sum, there is weak evidence that the sunshine rule will produce more frequent plaintiff awards than the blindfold rule, especially when the plaintiffs' perceived fault is close to fifty percent(n192)

H6: The mean dollar recovery of the plaintiff from all juries in the sunshine sample will be larger than from those in the blindfold sample.

Rationale: If, as we hypothesized in H5, the sunshine plaintiff will recover an ultimate award more often, based on that factor alone, the mean sunshine recovery should be larger than the mean blindfold recovery.

ER6: The mean ultimate recovery for the sunshine plaintiff was $1,029,186;(n193) for the blindfold plaintiff, it was $973,622.(n194) Based on the t-test difference of means test (t=.15, d.f.=80, p=.44), there is no statistically significant difference between these numbers. H6 is not supported by the data. The explanation appears to be that, although the sunshine plaintiff wins an ultimate award more often, that award tends to be smaller. Table 6 indicates the verdict results from the Base blindfold juries that made an award to Dawn. The average award for these twelve blindfold juries was $3,164,273.(n195) From Table 7, the average award made by the seventeen Variant sunshine juries was $2,606,234.(n196) Assuming that the same number of cases would be filed and decided under either rule, based on these data, the total payout to plaintiffs would be roughly the same under either rule. It is arguable that a compensation system that produces more frequent relief, even if it is less complete relief-- such as we see under the sunshine condition -- is preferable to one that has the lottery aspects of much of the present tort litigation system. No-fault systems like workers' compensation are based on this proposition.

SUMMING UP THE EXPERIMENTAL RESULTS

The mean verdict data strongly support the proposition that jurors who are aware of a percentage bar to recovery will react in ways generally perceived to be more favorable to the plaintiff than will jurors who are not privy to that information,(n197) Sunshine juries tend to increase the fault percentages of the defendants, but may do so disproportionately if so guided by the evidence and the attorneys' arguments. There is initial but weak evidence that sunshine juries will make ultimate awards more often to negligent plaintiffs, but even those juries that do not make an award tend to attribute lower percentages of fault to plaintiffs than do blindfold juries. The ultimate awards made by sunshine juries tend to be smaller, so that even with increased frequency, average recovery, and presumably total recovery as well, will be about the same

under sunshine and blindfold rules. Although it would seem that something like sympathy or dissatisfaction with the modified comparative fault concept has an effect on jury verdicts, the net economic impact of adopting either a sunshine or a blindfold rule for a jurisdiction appears to be statistically insignificant. This conclusion is reinforced by the fact that, when the plaintiff's fault is either slight or gross, the choice of whether to inform the jury about the percentage bar is likely to be immaterial.

The Economic Effects of Four Defensive Rule Regimes

The verdict data from this experiment provide an excellent opportunity to compare the economic effect four negligence defense rules would have on a typical case.

Contributory Negligence as a Complete Defense

Every one of the juries that rendered verdicts in McKey v. Torino Pizza found the plaintiff to be partially at fault. Presumably, under a contributory negligence rule, Dawn would recover nothing in every instance. We have read, however, that juries under this rule in the past have made ultimate awards to plaintiffs in derogation of the instructions and their findings.(n198) It would be interesting to test empirically how often that would happen with McKey juries charged under a contributory negligence instruction.

Pure Comparative Fault

The blindfold juries, whose verdicts are tabulated in Table 1, received what was essentially a pure comparative fault instruction. If pure comparative fault were the rule in Indiana, the mean recovery for Dawn would be $1,414,629.(n199)

Modified Comparative Fault Under a Blindfold Rule

If the Table 1 juries' verdicts are adjusted for modified comparative fault with a 51% percentage bar to recovery and the juries are instructed under a blindfold rule as this experiment assumes, and as Table 2 illustrates, Dawn's mean recovery would be $973,622.(n200) There seems little doubt that modified comparative fault favors defendants significantly more than does the pure system.

Modified Comparative Fault Under a Sunshine Rule

If we assume that the verdict results found in Table 3 are from juries similar to those tabulated in Tables 1 and 2, we see that the mean award under modified comparative fault where the jury learns of the 51% recovery bar would be $1,029,186.(n201) As noted, the aggregate economic difference between applying and lifting the blindfold in modified comparative fault cases appears to be statistically insignificant.(n202)

The Generalizability Question

In the behavioral sciences, laboratory experiments are vulnerable to the criticism that they lack realism, and as a result generalization may not be justified. But as Ilgen points out:

It is well accepted that all research, regardless of setting, requires trade-offs .... Thus all settings, whether laboratory, field, or some combination of the two, create contextual conditions that have both advantages and disadvantages for contributing to knowledge that generalizes to human behavior in ongoing organizations. For example, the naturalness of the field setting is purchased at the cost of control. Without some control, it is often impossible to disentangle the effects of many different covarying and confounded variables on the behaviors of interest. Is it better to obtain more realism going to the field yet sacrificing control, or is it better to gain control in the laboratory but lose some of the naturalism?(n203)

"The obvious answer," he says, "is that it depends."(n204)

Ilgen concludes that there are four sets of conditions that are relevant to the use of laboratory research: 1) when high fidelity can be established between laboratory and field; 2) when "laboratory conditions are to be recreated in the field"; 3) when "field conditions limit the feasibility of field research"; and 4) when "the hypothesis of interest is one demanding an effect rather than direct generalization of that effect to a particular setting."(n205)

In this experiment, we took pains to maximize the fidelity of the lab setting to the field setting of a real trial. First, we chose a real trial for the simulation models.(n206) We excerpted the testimony, exhibits, arguments, and instructions from the actual trial transcript. We built a realistic courtroom set and cast lawyers in the roles of the judge and counsel for plaintiff and defendant. The jurors observed a four hour simulation, long enough to develop the issues in a realistic manner, and most important, the jurors deliberated in panels similar in size to the actual juries that hear these cases. When asked on the exit survey whether their jury panel's deliberations were conducted as if this were a real trial; whether a real jury likely would have reached approximately the same result; whether they were satisfied with their panel's verdict; whether their panel's verdict was fair; and whether the process in reaching it was fair; overwhelming majorities of the individual jurors agreed or strongly agreed.(n207) Ilgen's condition of high fidelity was met.

Ilgen's second condition applies in the sense that the lab experiment seeks to test instructional rules that could be implemented in the field. Just as it makes sense to test prototype machinery in the lab before deploying units in the field, it makes sense to test blindfold and sunshine rules in the lab before enacting them as legislation.

Ilgen's third condition involves field constraints, such as time, expense, ethics, threats to health and safety, and the impossibility of field research. Lab research permits time compression consistent with fidelity imperatives. Studying the events of real trials would be hugely time consuming. Field research, if it were possible, would be prohibitively expensive. Both ethical and legal constraints impinge on the ability of field researchers to study what goes on in the jury room. And because of "the covarying and confounded variables on the behaviors of interest,"(n208) replicating lab research in the field is really not feasible, because of insufficient control.

The fourth condition under which lab research may be indicated is when the researchers are seeking to determine whether an effect can happen, rather than that it will happen under field conditions. We have shown that manipulating an independent variable by replacing a blindfold rule with a sunshine rule can have an effect on student mock juror subjects. Although we can not say with certainty that real jurors under field conditions will react the same, we think the evidence from our experiment is persuasive. There is also substantial empirical evidence that lab research using student subjects does generalize to field settings.(n209) While the dollar awards and the magnitude of differences may not be the same in a real world setting as is reported in this study, it is likely that the same general findings would be reached. We would urge that the external validity of our study be tested by replication.

CONCLUSION

From 1931 to the middle 1970s, every state adopting the modified form of comparative fault also opted to blindfold their civil juries with respect to the existence of a fifty or fifty-one percent bar to the plaintiff's recovery. In the late '70s, however, there was a dramatic turnabout: every new adopter of modification, and several that had earlier opted for blindfolding, either mandated or permitted this percentage bar information to be disclosed to the jury. That is, until 1995, when the Illinois General Assembly reverted to blindfolding Illinois juries under the rubric of "tort reform." The Assembly apparently concluded that blindfolding would benefit the state's economy by reducing the liability exposure of businesses, professionals, governmental entities, and other constituencies often found to be defendants in tort cases. It is likely that other states will be persuaded by this reasoning to follow Illinois's lead despite the lack of empirical evidence to support it.

In this article, the authors chronicle and analyze these events. While we take no political or philosophical position for or against jury blindfolding, we argue that legislation designed to produce an aggregate economic effect should, if at all possible, be empirically tested, preferably before enactment. To that end, we designed and executed a lab experiment to compare modified comparative fault verdicts rendered under blindfold rules against those rendered under sunshine rules. The data indicate that civil juries respond to sunshine rules by lowering the percentage of fault attributable to plaintiffs, but they also appear to temper this generosity by making smaller awards. While there is weak evidence that sunshine plaintiffs, under facts similar to those of the experiment, will recover damages more frequently, their mean dollar recovery will be no greater than under the blindfold rule. Thus, it can be argued that the aggregate effect of the two rules on the economy of the jurisdiction will be about the same.

Tort reform proponents will assert that, if a rule produces more frequent plaintiff victories, it will stimulate litigation and its attendant costs. We respond that smaller victories inhibit litigation(n210) so that the net economic effect of implementing sunshine rules is likely to be negligible. To the extent that our student jury verdicts are generalizable(n211)--and we believe we have made the case that our evidence is probative--the reimplementation of blindfold rules with respect to the percentage of negligence question will prove to be ineffective tort reform.

UPDATE

On December 18, 1997, the Illinois Supreme Court declared several core provisions of the 1995 Civil Justice Reform Act to be unconstitutional and the Act as a whole to be "void in its entirety."(n212) The principal target of the court's holding was the statute's $500,000 limitation on non-economic damages.(n213) With respect to the section on "[j]ury instructions in tort actions,"(n214) of which the blindfold rule on the percentage of negligence was a part, the court declared this rule and the one requiring disclosure of the non-taxability of damages to be "not clearly invalid."(n215) But because the invalidity of the damages cap and several other provisions left only "a residue" of what was supposed to be a comprehensive legislative initiative, the court refused to sever the valid provisions from the invalid.(n216) At this writing it is uncertain whether even the valid sections will be reenacted.(n217)

The supreme court discussed three scholarly affidavits submitted by the plaintiffs to challenge the eighteen findings placed in the preamble of the Act, upon which the Illinois General Assembly had relied to justify its tort reform initiative.(n218) The affidavits summarized a number of empirical field studies that found little or no adverse economic impact traceable to the type of civil litigation system in effect prior to the enactment of the Civil Justice Reform Act.(n219) The plaintiffs argued that, based on these studies, there could be :no rational basis for the legislation.(n220) The empirical evidence relied on by the legislature, they charged, had been "chiefly anecdotal."(n221)

The court stated it was powerless to "`adjudicate' the accuracy of legislative findings .... [Its] task is limited to determining whether the challenged legislation is constitutional, and not whether it is wise."(n222) But by simply discussing the plaintiffs' empirical evidence, the supreme court appeared to be declaring that legislatures have a duty to base legislation on sound factual findings, even though the findings may not be subject to judicial review. When the Illinois General Assembly meets to reconsider the rule against disclosing to civil juries the existence of a percentage bar to recovery, it is to be hoped it will seek out valid empirical studies--including those that employ the laboratory methods of social psychology--to guide its legislative course of action.

The authors gratefully acknowledge the invaluable assistance of the following:

Funding support: The Academy of Legal Studies in Business (ALSB); the J.I. Holcomb Research Institute; The Ind. U. Purdue U. in Indianapolis (IUPUI) Research Fellowship Program; and the AMOCO Foundation Multidisciplinary Research Grant Program. We especially wish to acknowledge the initiative of Professor O. Lee Reed, who as ALSB President proposed the award of a seed money grant for empirical research, without which this project could not have commenced.

Volunteer "actors" for the trial simulations: Mary Chappell, Joan Leibman, Christopher Dull, Thomas Weinland; Professors Robert Hall, Charles Luckenbill, Richard Magjuka, Roger Schmenner, and Jerrold Stern. Also, Michael Maitzen, director of the production for IUPUI's ComTechLab.

Academy of Legal Studies in Business faculty members who administered iterations of the trial simulations in which their students participated as mock jurors: Professors Caryn

Beck-Dudley, Utah State U.; Robert Bennett, Butler U., Don Boren, Bowling Green U.; Laura Ginger, Ind. U. (Bloomington); Jordan Leibman, Ind. U. (Indianapolis); Nim Razook, U. Okla.; Eric Richards, Ind. U. (Bloomington); Ellen Rubert, College of Lake County; Robert Van Der Velde, Auburn U. (Montgomery); Nancy Wainwright, Eastern Wash. U.; Dexter R. Woods, Ohio Northern U.; and Floyd Woodward, Ind. U. (Indianapolis).

Assistance with the project's statistical methodology: Professors Ronald Anderson, John Hassell, and Debra Mesch.

(n1) See infra notes 72-74, 94-96 and accompanying text.

(n2) This article focuses on rules that prohibit, require, or permit disclosure of the existence of a percentage bar to the plaintiffs recovery under modified comparative fault. Currently, only four modified comparative fault states blindfold on this issue: Illinois, 735 ILL. COMP. STAT. Sections 5/2-1107.1 (1996) (reprinted in its entirety infra note 12); Ohio, McClure v. Neuman, 178 N.E.2d 621 (Ohio Ct. App. 1961); Texas, TEX. R. CIV. P. 277; and Wisconsin, McGowan v. Story, 234 N.W.2d 325 (Wis. 1975).

Some states prohibit, require, or permit disclosure of other legal consequences to civil juries such as the question of who pays if a defendant is insolvent, immune or unavailable for suit, or if a nonparty is found to be partially at fault in causing the harm complained of. E.g., COLO. REV. STAT. Sections 13-21-111.5(5) (1996) provides that "the jury shall not be informed as to the effect of its finding as to the allocation of fault among two or more defendants." WYO. STAT. Sections 1-1-109 (1996) provides that the trial court shall "[i]nform the jury. of the consequences of its determination of the percentage of fault," which has been interpreted to mean that the jury must be informed as to the 51% percentage bar to recovery; that the court will reduce the award by the plaintiffs percentage of fault; that each defendant will be liable only to the extent of its percentage of fault (no joint and several liability); and that the plaintiff cannot recover for negligence attributed to a nonparty. However, the jury is not to be informed when a defendant is insolvent. John M. Burman, Wyoming's New Comparative Fault Statute, 31 LAND & WATER L. REV. 509, 529-30 (1996), citing Wyo. Civ. Pattern Jury Instruc. Nos. 10.03, 10.05 (1994) and Burton v. Fisher Controls Co., 723 P.2d 1214, 1222 n.6 (Wyo. 1986). The Illinois statute forbids the court to disclose to the jury. limitations on non-economic and punitive damages, but requires the court to inform the jury that verdict awards will be non-taxable. 735 ILL. COMP. STAT. Sections 5/2-1107.1 (1996), See infra notes 152-57 and accompanying text.

(n3) For an excellent discussion of jury nullification, see JEFFREY ABRAHAMSON, WE THE JURY 57-95 and references 266-72 (1994), The author describes the 1735 trial of John Peter Zenger for the "infamous crime of seditious libel," a law that made it criminal to publish "written censure upon any public man whatsoever, or upon any law or institution whatsoever." ABRAHAMSON, supra, at 73-74. The jury ignored the court's instruction to limit its findings to whether the defendant had published the clearly censorious material and it acquitted Zenger. Abrahamson refers to the Zenger case as "the defining moment for the American jury in the colonies." Id. at 73-74. The author tells us that "[w]ell into the nineteenth century, criminal juries frequently (and civil juries occasionally) were instructed that the judge's statement of law was not binding on them; that they could determine for themselves what the law was." Id. at 63-64

(citations omitted). Jury nullification is generally discussed in the context of the criminal law where the power to acquit is bolstered by the double jeopardy doctrine. Civil verdicts are subject to the trial court's power to order remittitur, and the appellate court's power to reverse large judgments on the ground that the verdicts must have been motivated by passion or prejudice sufficient to shock the court's conscience. See also Nancy J. King, Juror Delinquency in Criminal Trials in America, 1776-1996, 99 MICH. L. REV. 2673, 2736-37 (1996) (speculating that jury nullification may be the wave of the future).

(n4) ABRAHAMSON, supra note 3, at 91-92.

(n5) We define "sunshine rule" as a rule requiring or permitting the legal consequences of a jury finding to be made open and transparent to the jury. We analogize here to the term "sunshine law," a law mandating that meetings of governmental bodies be open to the public. See, e.g., James E. Costello, Comment, The Limits of Popular Sovereignty: Using the Initiative Power to Control Legislative Procedure, 74 CAL. L. REV. 491, 514 n. 119 (1986). In contrast, courts and commentators have used the term "blindfold" to characterize rules that bar disclosing to juries the legal consequences of their findings. See, e.g., Kaeo v. Davis, 719 P.2d 387, 395 (Haw. 1986); Adkins v. Whitten, 297 S.E.2d 881, 882 (W. Va. 1982); Moore v. Swoboda, 571 N.E.2d 1056, 1063 (Ill. Ct. App 1991); Leon Green, Blindfolding the Jury, 33 TEX. L. REV. 273 (1955); Michael J. Norton, Comment, McGinn v. Utah Power & Light Co. -- Jury Blindfolding in Comparative Negligence Cases, 1975 UTAH L. REV. 569.

Sunshine rules can be "mandated," see, e.g., COLO. REV. STAT, ANN. 13-21-111.5(5) (West Supp. 1995) ("[T]he court shall instruct the jury.., as to the degree of negligence .... "); "requested," see, e.g., Adkins v. Whitten, 297 N.E.2d 881, 884 (W. Va. 1982) (court has duty to inform jury when requested); or "discretionary," see, e.g., Roman v. Mitchell, 413 A.2d 322, 327 (N.J. 1980) (court should inform jury of effect of answers to interrogatories, but has the discretion to withhold).

(n6) For lists and analysis of the authorities for these provisions, see VICTOR E. SCHWARTZ, COMPARATIVE NEGLIGENCE, Appendix B, at 391-419 (2d ed. 1986 and Supp. 1993) [hereinafter SCHWARTZ]; HENRY WOODS & BETH DEERE, COMPARATIVE FAULT 22-26, Appendix 515-932 (3d ed. 1996) [hereinafter WOODS].

The following modified comparative fault states set their percentage bar to recovery at 50%; in these states, a finding of equal fault on the part of plaintiff and defendant means that the plaintiff is barred from all recovery: Ariz., Ark., Colo., Ga., Idaho, Kan., Maine, Neb., N.D., Tenn., Utah, W. Va. WOODS, supra Sections 1:11, at 24-25. This rule is sometimes referred to as the Georgia plan, after the first state to apply the rule. Id. See infra notes 81-83 and accompanying text for a discussion of why this formula is particularly problematic under a blindfold rule.

The following modified comparative fault states set their percentage bar at 51%; in these states a plaintiff whose fault is equal to that of the defendant recovers 50% of her damages: Conn., Del., Haw.*, Ill.*, Ind., Iowa*, Mass.*, Minn.*, Mont., Nev., N.H., N.J., Ohio, Okla.*, Ore.*, Pa., S.C., Tex., Vt., Wis.*, Wyo.*. This rule is referred to as the New Hampshire plan. States marked

with an asterisk have switched from the Georgia to the New Hampshire plan. WOODS, supra, Sections 1:11, at 23-24.

In three states, Wis., Minn., and Idaho, if there are multiple defendants, the plaintiffs fault must be compared with each defendant's separately so that where plaintiff "is 30% negligent and defendants D, E, F are 25%, 25%, and 20% negligent respectively, plaintiff recovers nothing." WOODS, supra., Sections 13:1, at 226.

(n7) For cases and commentary generally in favor of blindfold rules, see McClure v. Neuman, 178 N.E.2d 621, 624 (Ohio Ct. App. 1961) ("Where the court indicates how liability will be determined, it warns the jury of the result of its answers and permits the jury to `trim its course accordingly.'") (quoting 59 Ohio Opinions 381); McGowan v. Story, 234 N.W.2d 325 (Wis. 1975) ("[I]t is reversible error for either the court or counsel to inform the jury of the effect of their answer on the ultimate result of their verdict .... "). See also infra notes 66-85 and accompanying text for discussion of the Wisconsin experience; James G. Denton, Informing a Jury of the Legal Effect of its Answers, 2 ST. MARY'S L.J. 1, 10 (1970) (arguing that in jurisdictions using a "special issues" (special verdict questions) system, the system's purposes "would best be served by limiting the jury's considerations to the facts and not their consequences"); Coleman Gay, Blindfolding the Jury: Another View, 34 TEX. L. REV. 368, 380 (1956) ("If juries are to be allowed to disregard the court's instructions and to decide for one party or the other regardless of the evidence, the parties' rights will depend solely on the conscience of the particular jury trying the case."); Stuart F. Schaffer, Comment, Informing the Jury of the Legal Effect of Special Verdict Answers in Comparative Negligence Actions, 1975 DUKE L.J. 824, 851 (concluding that informing the jury where special verdicts are used "violates the comparative negligence scheme established by the legislature or the highest court of a state .... "). For authorities from pure comparative fault states favoring blindfolding juries from information other than the existence of a percentage bar, see Robinson v. Murlin Phillips & MFA Ins. Co. 557 S.W.2d 202, 203 (Ky. 1977) ("In order to [decide contested issues of fact] there is no need for jurors to know the legal effect of their resolution .... "); Mitchell v. Perkins, 54 N.W.2d 293, 296 (Mich. 1952) (holding that, in Michigan, a general verdict state, the jury must not be told in advance what facts are necessary to support the verdict).

For cases and commentary generally opposed to blindfold rules, see Kaeo v. Davis, 719 P.2d 387 (Haw. 1986) (arguing that attempts to keep the jury in the dark will be unavailing); Seppi v. Betty, 579 P.2d 683 (Idaho 1978) (pointing out the danger of failing to inform the jury in states where a 50-50 finding of equal fault bars the plaintiff from recovering); Moore v. Swoboda, 571 N.E.2d 1056 (Ill. Ct. App. 1991) ("The failure to inform the jury of the effect of its findings in such circumstances [one percentage point difference can mean no recovery] may result in a decision shaped by a misimpression of the law."); Roman v. Mitchell, 413 A.2d 322 (1980) (observing that, before comparative fault, jurors were traditionally informed of the legal effect of a contributory negligence finding); Adkins v. Whitten, 297 S.E.2d 881 (W. Va. 1982) ("[U]nder our jury trial system, it is incumbent on the court., o to inform the jury as to the law that is applicable to the facts of the case."); Green, supra note 5, at 282 ("[T]here is no blindfold known that will prevent a jury from thinking they know the legal effect of their answers."); Glenn E Smith, Comparative Negligence Problems with the Special Verdict: Informing the Jury of the Legal Effects of Their Answers, 10 LAND & WATER L. REV. 199, 226-28 (1975) (arguing that

the blindfold rule leads to erroneous suppositions by juries seeking an ultimate outcome); Note, Informing the Jury of the Effect of its Answers to Special Verdict Questions -- The Minnesota Experience, 58 MINN. L. REV. 903, 927 (1974) (arguing that the Minnesota Supreme Court's attempt to limit disclosure only to comparative negligence cases was misguided because the rule "does not consider the role of the common sense wisdom of juries in mitigating unfair laws and producing just results in individual cases"); Norton, supra note 5, at 581 ("[A] continued parroting of unfounded concern over jury sympathy and bias [will] result in nothing more than the preservation of a useless and even counterproductive doctrine.").

(n8) The comparative negligence amendments were part of The Civil Justice Reform Amendments of 1995, Pub. Act No. 89-7, 1995 Ill. Legis. Serv. 224 (West) (codified in scattered sections of ILL. REV. STAT. chs. 430, 730, 735, 740, 745, 815, 820 (West Supp. 1996)).

(n9) See generally Kirk W. Dillard, Illinois' Landmark Tort Reform: The Sponsor's Policy Explanation, 27 LOY. U. CHI. L.J. 803, 816 (1996). For a discussion of the Illinois experience, see infra notes 116-57 and accompanying text.

(n10) Dillard, supra note 9, at 816 (1996).

(n11)See infra note 58 for a list of the 33 states currently maintaining modified comparative fault systems and the source of each.

(n12) Former ILL. REV. STAT. ch. 110, par. 2-1107.1 was converted to new 735 ILL. COMP.

STAT. Sections 5/2-1107.1 (West 1996) as follows: Jury instruction in tort actions. In all actions on account of bodily injury or death or physical damage to property based on negligence, or product liability based on [any theory or doctrine, strict tort liability, the court shall instruct the jury in writing, to the extent it is true, that any award of compensatory damages or punitive damages will not be taxable under federal or State income tax law. The court shall not inform or instruct the jury] that the defendants shall be, found not liable if the jury finds that the contributory fault of the plaintiff is more than 50% of the proximate cause of the injury or damage for which recovery is sought, [but it shall be the duty of the court to deny recovery if the jury finds that the plaintiff's contributory fault is more than 50% of the proximate cause of the injury or damage. The court shall not inform or instruct the jury concerning any limitations in the amount of non-economic damages or punitive damages that are recoverable, but it shall be the duty of the trial court upon entering judgment to reduce any award in excess of such limitation to no more than the proper limitation.]

[This amendatory Act of 1995 applies to causes of action filed on or after its effective date]. The Illinois Civil Justice Reform Act of 1995 (sometimes referred to as the Illinois Tort Reform Act) is reproduced with strikeouts from, and additions to, the old statute at 27 LOY. U. CHI. L.J. 819 (1996).

(n13) See infra notes 21-65 and accompanying text.

(n14) See infra notes 66-157 and accompanying text.

(n15) See infra notes 177-80, 188-92 and accompanying text.

(n16) See infra notes 193-96 and accompanying text.

(n17) See infra note 198 and accompanying text.

(n18) See infra note 199 and accompanying text.

(n19) See infra note 200 and accompanying text.

(n20) See infra notes 201-02 and accompanying text.

(n21) See, e.g., ARK. CODE ANN., Section 16-64-122(c) (1997) ("The word fault as used in this section includes any act, omission, conduct, risk assumed, breach of warranty, or breach of any legal duty which is a proximate cause of any damages sustained by any party."); 735 ILL. COMP. STAT. Section 5/2 1116(b) (West 1996) ("Fault means any act or omission that (i) is negligent, willful and wanton, or reckless, is a breach of an express or implied warranty, gives rise to strict liability in tort, or gives rise to liability under the provisions of any state statute, rule, or local ordinance .... "); IND CODE Section 34-4-33-2(a)(1) (1997) ("`Fault' includes any act or omission that is negligent, willful, wanton, reckless or intentional toward the person or property of others .... ") (emphasis added).

With respect to strict liability, in the leading case, Daly v. General Motors Corp, 575 P.2d 1162 (Cal. 1978), the court held that, while there were "theoretical and semantic distinctions between the twin principles of strict product liability and traditional . . . negligence they can be blended and accommodated." Id. at 1167. For a discussion of the role of strict liability in the transformation of comparative negligence into comparative fault, see Jordan H. Leibman, Comparative Contribution and Intentional Torts: A Remaining Roadblock to Damages Apportionment, 30 AM. BUS. L.J. 678, 684-87 (1992).

(n22) Early federal and state legislation applied comparative negligence principles to cases involving injured workers, but in 1910, Mississippi enacted a "pure" form statute that applied to all negligence actions. See SCHWARTZ, supra note 6, Section 1.4(B), at 11-12; WOODS, supra note 6, Section 1:11, at 19-20.

(n23) See W. PAGE KEETON ET AL., PROSSER AND KEETON ON THE LAW OF TORTS Section 65, at 451 n.1 (5th ed. 1984) [hereinafter PROSSER & KEETON] ("The earliest contributory negligence case is Butterfield v. Forrester, 1809, 11 East 60, 193 Eng. Rep. 926.").

(n24) "The first American case appears to have been Smith w Smith, 1824, 19 Mass. (2 Pick) 621." Id.; See also WOODS, supra note 6, Section 1:3, at 5-6.

(n25) SCHWARTZ, supra note 6, Section 1.3, at 8; WOODS, supra note 6, Section 1:9, at 13-14.

(n26) See Ernest A. Turk, Comparative Negligence on the March, 28 CHI.-KENT L REV. 189, 220-25 (1950).

(n27) See William Prosser, Comparative Negligence, 41 CAL. L. REV. 1, 1-4 (1953). `The last vestiges of the complete defense disappeared long since from continental Europe, which divides the damages, Great Britain, all of the Canadian provinces, New Zealand... have come to the same result so that little of the British Empire is left with the common law rule." Id. at 2; WOODS, supra note 6, at 14.

(n28) See SCHWARTZ, supra note 6, Section 1.3(B), at 9-10; WOODS, supra note 6, Section 1:10, at 14-17.

(n29) Prosser, supra note 27, at 4.

(n30) See PROSSER AND KEETON, supra note 23, Section 65, at 452-53. The authors suggest that the defense has "a penal basis in which the plaintiff is punished for his own misconduct"; or it may derive from the equitable "clean hands" doctrine, or based on a proximate cause analysis, the plaintiffs negligence may be seen as "an intervening, or insulating cause" of her injury. The concept of contributory negligence tracks with the tendency of early 19th century courts to look for "a single, principal, dominant, `proximate' cause of every injury." The complete defense's harsh bar to recovery is said to discourage accidents and to represent "the highly individualistic attitude of the common law" reflecting "an uneasy distrust of the plaintiff-minded jury . . . in the early 19th century." Finally, it accommodated "a desire to keep the liabilities of growing industries within some bounds."

This last rationale is explored in MORTON J. HOROWITZ, THE TRANSFORMATION OF AMERICAN LAW 1780-1860 (1977). Horowitz describes how American courts in the last two thirds of the 19th century profoundly restructured common law rules to accommodate economic development. Laissez faire ideas transformed the common law of contracts while liability-limiting rules dominated the evolution of tort law. The emergence of modern negligence principles to replace the causation-based rules of trespass was the principal engine for this accommodation because negligence theory required plaintiffs to prove fault in addition to causation.

(n31) See WOODS, supra note 6, Section 1:15, at 8-9.

(n32) See WOODS, supra note 6, Sub-Section 1:16-17, at 9-11. The "last clear chance" doctrine negates the contributory negligence defense when the defendant is aware (or in some states, should be aware) that he can prevent injury to a negligent plaintiff and fails to do so. Judge Woods notes that "[a] number of commentators have observed that last clear chance is one step, and a rather significant one, toward a system of comparative negligence." Id. Section 1:7, at 12.

Among the limitations imposed by courts in addition to "last clear chance" are the requirement that it would not be contributory negligence if the plaintiffs were not warned of specific hazards; or if the plaintiff were acting in an emergency to save human life or valuable property; or when the plaintiffs attention was diverted by the defendant's signs; or when the defendant has acted intentionally, recklessly, or wantonly; or when the defendant's act was subject to absolute liability. Id. Section 1:6, at 9-10.

(n33) For a discussion of the Illinois experience, see infra notes 116-57 and accompanying text.

(n34) See WOODS, supra note 6, at 601 (citing the statute now appearing at GA. CODE ANN. Section 46-8-291).

(n35) 79 S.E. 836 (Ga, 1913).

(n36) Id. (quoting the statute now appearing at GA. CODE ANN, Section 51-11-7).

(n37) Christian v. Macon Ry. & Light Co., 47 S.E. 923 (Ga. 1904).

(n38) Id.

(n39) Act of April 22, 1908, oh. 149, Section 3, 35 Stat. 66, 45 U.S.C. Section 53.

(n40) SCHWARTZ, supra note 6, Section 1.4(A), at 11. Schwartz notes that, in 1920, Congress incorporated the same principles into the Jones Act, Act of June 5, 1920, oh. 250, Section 33, Stat. 1007, 46 U.S.C. Section 688 (protecting seamen suffering injury or death); and the Death on the High Seas Act, Act of Mar. 30, 1920, ch. 111, Section 6, 41 Stat. 537, 46 U.S.C. Section 766. Id.

(n41) Id.

(n42) Currently codified at MISS. CODE ANN. Section 11-7-15 (1997).

(n43) SCHWARTZ, supra note 6, Section 1.4(B), at 11-12.

(n44) Id. at 12. For a discussion of the Wisconsin experience, see infra notes 66-85 and accompanying text.

(n45) See infra note 76 and accompanying text.

(n46) "The influence of Wisconsin on the adoption of [comparative negligence] and its interpretation has been enormous." WOODS, supra note 6, at 920.

(n47) Id. at 12.

(n48) WOODS, supra note 6, Section 1:11, at 21-22.

(n49) Id. at 20-21. Arkansas's first statute provided for the pure form, and it required mandatory special verdicts. The stats's insurance bar was displeased with the pure form and the plaintiffs bar was unhappy because it found rural jurors were unable to understand the special verdict questions. The statute was replaced in 1957 with a modified statute having a 50% recovery bar and no mandatory special verdicts. Id.

(n50) See SCHWARTZ, supra note 6, Section 1.4(B), at 12-13.

(n51) Henningsen v. Bloomfield Motors, 161 A.2d 69 (N.J. 1960), the case felling the citadel of privity in warranty cases thus opening the door to the much wider use of breach of implied warranty as a quasi-tort strict liability theory, and Greenman v. Yuba Power Products, Inc., 27 P.2d 897 (Cal. 1962), the case establishing and articulating the policies underpinning strict liability in tort for defective products (Section 402A liability) were both decided during the early part of the 1960s. See PROSSER AND KEETON, supra note 23, Section 97, at 690-94. "Section 402A liability in tort swept the country, just as the expansion of warranty liability under Henningsen had done .... "Id. at 694.

(n52) SCHWARTZ, supra note 6, Section 1.4(B), at 13-14 & n.39.

(n53) Id. at 13-14.

(n54) See id. Section 1.6, at 26 ("Reform of the fault system appears to be a compromise between pressures for `no-fault' systems on one side and for the status quo on the other. Because comparative negligence is an important part of this compromise, it is likely to become the law in all states in the near future.").

(n55) 280 So. 2d 431 (Fla. 1973).

(n56) The following states judicially adopted and currently maintain the pure form of comparative fault: Alaska, Cal., Fla., Ken., Mich., Mo., N.M. WOODS, supra note 6, Section 1:11, at 22-23. W. Va., S.C., Tenn., judicially adopted modified systems, Id. at 22. Illinois initially adopted judicially the pure form but later switched legislatively to a modified system. See infra notes 14346 and accompanying text.

(n57) In addition to the seven judicial-adoption, pure form states listed supra note 56, the following have enacted the pure form: La., Miss., N.Y., R.I., Wash. WOODS, supra note 6, Section 1:11, at 22-23. Arkansas originally enacted a pure form statute, but switched to the modified form. See supra note 49 and accompanying text.

(n58) The following are the 33 states that have enacted or have judicially adopted modified comparative fault; the source of each is indicated: ARIZ. REV. STAT. Sub-Section 12-2501, 12-2505, 12-2506, 12-2508, and 12-2509 (1993); ARK. CODE ANN. Sub-Section 16-64-122 (1991); COLO. REV. STAT. Section 13-21-111.5 and 13-21-111.7 (1973 & Supp. 1978); CONN. GEN. STAT. ANN. Section 52572h (West Supp. 1980); DEL. CODE tit. 10, 8 8132 (1984); GA. CODE ANN. Sub-Section 51-12-31:, 1512-32, 51-12-33 (1987); HAW. REV. STAT. Section 663-10.9 (1991); IDAHO CODE Sub-Section 6-801 to 6-806 (1987); 735 ILL. COMP. STAT. Section 5/2 1107.1, 5/2-1116, 5/2-1117 (West 1996); IND. CODE 88 34-4-33-1 to 34-4-33-12 (1995); IOWA CODE ANN. Section 619.17 Sub-Section 668.1 to 668.10 (1984); KAN. STAT. ANN. Section60-258a and 60-258b (1976); ME. REV. STAT. ANN. tit. 14, 8 156 (1980); MASS. GEN. LAWS ANN. ch. 231, Section 85 (Supp. 1978); MINN. STAT. ANN. Sub-Section 604.01 and 604.01 (Supp. 1981); MONT. CODE ANN. Sub-Section27-1-719(5) 27-1-702, 27-1-703 (1987); NEB. REV. STAT. Sub-Section 2521,185 through 25-21,185.12 (1992); NEV. REV. STAT. Section 41.141 (1987); N.H. REV. STAT. ANN. Sub-Section 507:7-d, 507:7-e (1990); N.J. STAT. ANN. Sub-Section 2A:15:5.1, 2A:15-5.2, 2A:15-5.3 (Supp. 1980-

81); N.D. CENT. CODE Section 32-03.2-02 (1993); OHIO REV. CODE Sub-Section 2315.19, 2315.20 (Anderson 1987); OKLA. STAT. ANN. tit. 23, Sub-Section 13 to 14 (West Supp. 1980-81); OR. REV. STAT. Sub-Section 18.470, 18.475 to 18.490; 818.510 (1987); PA. STAT. ANN. tit. 42, Section 7102(A) (Purdon Supp. 1980); McIntyre v. Balentine, 833 S.W.2d 52 (Tenn. 1992); Nelson v. Concrete Supply Co., 399 S.E.2d 783 (S.C. 1991); TEX. CIV PRAC. & REM. CODE ANN. Sub-Section 33.001, 33.002, 33.003, 33.011, 33.012, 33.013 (1987); UTAH CODE ANN. Sub-Section 78-27-37 to 7827-42 (1986); VT. STAT. ANN. tit. 12 8 1036 (1979); Bradley v. Appalachian Power Co., 256 S.E.2d 879 (W. Va. 1979); WIS. STAT. ANN. Section 895.045 (West Supp. 1980); WYO. STAT. Section 1-1109 (1995).

(n59) WOODS, supra note 6, 8 1:11 at 25.

(n60) Ala., Md., N.C., Va. Id.

(n61) See infra notes 199-202 and accompanying text.

(n62) "There is no better justification for allowing a defendant who is 49% at fault to escape liability than there is to allow a defendant who is 99% at fault under the old rule [contributory negligence] to escape liability." Alvis v. Ribar, 421 N.E.2d 886, 898 (Ill. 1981). See PROSSER & KEETON, supra note 23, Sections 67 at 473 n.42 (listing authorities for and against the modified system). The modified system of comparative fault has been described as "simply shift[ing] the lottery aspect of the contributory negligence rule to a different ground." Nga Li v. Yellow Cab Co., 119 Cal. Rptr. 858, 874 (Cal. 1975).

(n63) There is a split of authority whether each party's causal contribution to an injury should be compared as well as the relative fault of the parties. One view holds that "once causation is found, the apportionment must be made on the basis of comparative fault rather than contribution." William Presser, Comparative Negligence, 51 MICH. L. REV. 465, 481 (1953). Other authorities argue that both fault and causal contribution should be compared. See WOODS, supra note 6, Sections 5:4, at 118-19 (discussing commentators and cases advancing this view). However the instruction is couched, it is clear that the jury will consider the quality of the acts proximately causing the injury in making their apportionment.

There is also the problem of apportioning causation. Once multiple causes have joined to produce an indivisible injury, it is not simple to develop a principled method for assigning causal portions to all antecedent actors. In the late 1970s, in several product liability cases, courts ostensibly apportioned damages using a theory of comparative causation, the most noteworthy being General Motors v. Hopkins, 548 S.W.2d 344 (Tex. 1977). The effect was to reduce the plaintiffs judgment rendered under strict liability in proportion to the causal contribution of the plaintiffs "unforeseeable misuse." The goal was to circumvent the "apples and oranges" problem of comparing negligence with strict liability. The concept is discussed in Aaron D. Twerski, The Many Faces of Misuse: An Inquiry into the Emerging Doctrine of Comparative Causation, 29 MERCER L. REV. 403 (1978).

Other writers have sought to develop a general theory of causal apportionment for tort cases. See Mario J. Rizzo & Frank S. Arnold, Causal Apportionment in the Law of Torts: An Economic

Theory, 80 COLUM. L. REV. 1398 (1980). These authors employ the concept of "probabilistic marginal product" (PMP) to develop a methodology for assigning damage shares in simultaneous cause cases "by measuring the differential degree of risk to which each cause exposes the plaintiff." Id. at 1408. This approach was criticized in David Kaye & Mikel Aicken, A Comment on Causal Apportionment, 13 J. LEGAL STUD. 219 (1986).

(n64) See infra notes 193-96 and accompanying text.

(n65) With respect to a percentage bar to recovery, the court may be required, forbidden, or permitted to disclose its existence, either upon a party's request or at the court's discretion. See supra note 5. The jurisdiction may also provide how the disclosure is to take place, how prior knowledge and logical inferences are to be handled, and whether counsel may argue the information to the jury. See, e.g., IND. CODE Sections 34-4-33.5(2) (1995): "If the percentage of fault of the claimant is greater than fifty percent (50%) of the total fault involved in the incident which caused the claimant's death, injury, or property damage, the jury shall return a verdict for the defendant and no further deliberation of the jury is required." Although the existence of the 51% bar is less than explicit, it is virtually certain the jury will infer it.

(n66) Although Wisconsin has relied heavily on the use of special verdicts over the years, the procedure has not been incorporated specifically into the state's comparative negligence statute. SCHWARTZ, supra note 6, Sections 17.4, at 302. Rather, the courts are to refer to the general special verdict statute codified at WIS. STAT. Sections 805.12 (1996). Subsection (1) provides:

Unless it orders otherwise, the court shall direct the jury to return a special verdict. The verdict shall be prepared by the court in the form of written questions relating only to material issues of ultimate fact and admitting a direct answer. The jury shall answer in writing. In cases founded upon negligence, the court need not submit separately any particular respect in which the party was allegedly negligent. The court may also direct the jury to find upon particular questions of fact.

See WOODS, supra note 6, at 921 (noting that the Judicial Council Committee note to the statute states that "[s]ubsection (1) is based on a recognition that in Wisconsin practice, the special verdict is the rule and not the exception").

(n67) A special verdict "is a special finding of facts of the case by a jury, leaving to the court the application of the law to the facts thus found" BLACK'S LAW DICTIONARY 1731 (4th ed. 1968). A general verdict is "[t]hat by which [the jury] pronounce[s] generally upon all or any of the issues either in favor of the plaintiff or defendant;--distinguished from a special verdict, which is that by which the jury finds facts only." Id.

A general verdict accompanied by special interrogatories is a hybrid form in which the jury answers specific fact questions but is also required to render a verdict for or against either party. See Nollenberger v. United Airlines, Inc., 335 F.2d 379, 405 n.41 (9th Cir.), cert. dismissed, 379 U.S. 951 (1964) (applying FED. R CIV. P. 49(b), which provides in pertinent part that the "court may submit to the jury, together with appropriate forms for a general verdict, written interrogateries upon one or more issues of fact the decision of which is necessary to a verdict").

Rule 49(a) provides that "[t]he court may require a jury to return only a special verdict in the form of a special written finding upon each issue of fact." FED. R. CIV. P. 49(a). Because the form of verdict is a matter of procedural law, the federal rules take precedence in diversity cases. Thus, even in a state like Wisconsin, where "special verdicts are the rule," see supra note 66 and accompanying text, a federal district court sitting in diversity could require a general verdict with interrogatories. Although Professor Schwartz supports special verdict practice in comparative fault cases, see infra note 69, he also looks with favor on the Rule 49(b) procedure: "A newer procedural device, jury interrogatories, captures most of the advantages of the special verdict procedures but avoids its disadvantages." SCHWARTZ, supra note 6, Sections 17.4, at 307.

(n68) Wisconsin's special verdict submissions in negligence cases require written answers to detailed questions asking the jury to determine whether the parties were negligent in the performance or omission of specific acts and whether those acts or omissions were causes of the injury complained of. Only if the jury finds negligent acts or omissions to have caused the injury is it to proceed to apportion negligence in percentages attributable to the parties. Finally, the jury is to determine the amount of damages of each type claimed. SCHWARTZ, supra note 6, Sections 17.4, at 302. For examples of typical special verdict forms used in Wisconsin courts, see id. at 302-05; Smith, supra note 7, at 215 n.56. For examples of recommended special verdict forms and general verdict forms accompanied by interrogatories, see WOODS, supra note 6, at 462-94.

(n69) For a brief discussion of the advantages and disadvantages of special verdicts generally and in comparative fault cases specifically, see SCHWARTZ, supra note 5, Sections 17.4, at 306-08. The author lists three principal advantages: 1) Control of the Jury: "The procedure causes the jury to separate in its mind the question of the amount of damages plaintiff suffered from the question of the percentage of fault." Id. at 307-08; 2) Facilitating Judicial Review: "Special verdicts (or jury interrogatories) localize error and allow a court to find the remaining portion of the verdict valid." Id. at 308; 3) Simplification of Instructions: If it is unnecessary to inform the jury of the legal consequences of its findings, the instructions can be greatly simplified. Id. at 308-09. The first two advantages also accrue to general verdicts accompanied by interrogatories, but not the third. As to jury control, the author acknowledges that critics of special verdict procedure find this to be a disadvantage because it denies the jury "`flexibility' in dealing with law that may not be generally popular." Id. at 308. Schwartz points out that "comparative negligence is of comparatively recent legislative judgment. Therefore, the law should be applied as a legislature intended it, or it should be changed at that level." Id. Professor Schwartz notes that another disadvantage of special verdicts is the difficulty in framing the questions: "By accident or design, the special verdict may fail to cover all the issues in the case or may contain incorrect terminology." Id at 306. Schwartz supports the use of special verdicts in comparative fault cases:

At first blush it is difficult to see why the application of a special verdict procedure to comparative negligence should be criticized. If the special verdicts are spelled out in the statute, the awkwardness that compromised the common-law special verdict system is eliminated. If they are not spelled out, recourse can be had to the useful forms developed in Wisconsin and other states.

Id. at 307. Another commentator differs: "More than thirty-five years of trial experience under both the pure and modified systems of comparative negligence, and having cases submitted on both general verdicts and interrogateries have convinced the writer that in many cases a general verdict is preferable." WOODS, supra note 6, Sections 18:1, at 439.

For a fascinating account of jury interrogatories gone awry, see JONATHAN HARR, A CIVIL ACTION 368-69 (1996) (describing the famous Woburn, Massachusetts environmental pollution case against Beatrice Foods and W.R. Grace, in which the lawyers and judge managed to produce four incredibly convoluted conditional questions for the jury; "All in all, the questions had the quality of a text that had been translated from English into Japanese and back again." Id. at 369).

(n70) Ryan v. Rockford Ins. Co., 46 N.W. 885, 886 (Wis. 1890).

(n71) See supra notes 39-42 and accompanying text.

(n72) Banderob v. Wisconsin Central Ry., 113 N.W. 738, 751 (Wis. 1907).

(n73) 1931 Wis. Laws 242 (current version codified at WIS. STAT. ANN. Sections 895.045 (West Supp. 1996)). See Schaffer, supra, note 7, at 830-31.

(n74) See supra note 66.

(n75) Pecor v. Home Indem. Co., 291 N.W. 313 (Wis. 1940).

(n76) See Schaffer, supra note 7, at 832 and nn.42-45 (citing the adoption of the Wisconsin (special verdict with blindfold) model by N.D., Tex., Kan., Ohio, Ark, Colo., Idaho, Utah, Wyo.).

(n77) Id. at 832-33.

(n78) For an excellent discussion of how jurors "expect to be involved, to a certain extent, in the application of the law" and "tend to dig for hidden legal meaning," see Norton, supra note 5, at 578-79.

(n79) Schaffer, supra note 7, at 833 (citing Banderob v. Wis. Cent. Ry., 111 N.W. 738 (Wis. 1907)). In 1907, contributory negligence was a complete defense in Wisconsin. It is entirely probable that juries inferred this rule despite special verdict procedures. It seems quite unlikely to the authors that juries would infer the 50% bar to recovery that was enacted in 1931. On the contrary, juries might easily and mistakenly infer instead the existence of a pure comparative negligence rule.

(n80) Id. at 834. Typically, a yes answer to an early question might imply potential liability, while a no would not. A Texas appellate court held that conditional questions are "permissible if the jury is not thereby informed of the effect of previous answers that determine the case." Id.

(citing Schroeder v. Rainboldt, 97 S.W.2d 679, 684 (Tex. 1936) (approving opinion of Comm'n App.)).

(n81) "In one situation, however, the jurors may conjecture incorrectly; they may believe that a plaintiff whose negligence is 50% of the total recovers 50% of his damages." SCHWARTZ, supra note 6, Sections 17.5, at 312.

(n82) Four reasons why juries tend to "split the difference" are given in Schaffer, supra note 7, at 831 n.35: 1) polarization of the jury; 2) the jury's inability to draw a finer distinction; 3) a credibility contest resulting from conflicting evidence; 4) the jury's realistic reflection of responsibility. Professor Schwartz comments that, "when the jury is not aware that a plaintiff who is 50% negligent recovers nothing, it may casually return a 50-50 verdict as a compromise." He is persuaded that the jury should be informed of the legal effect of the apportionment of fault .... at least in 50% systems." SCHWARTZ, supra note 6, Sections 17.5, at 312-13.

(n83) Schaffer, supra note 7, at 831-32. For a list of the states that also raised their percentage bar to 51%, see supra note 6.

(n84) Id. at 834-35. E.g., the Texas Supreme Court ruled that a charge would not be objectionable if it incidentally advised the jury of the effect of its answers. Id. (citing TEX. R. CIV. P. 277 promulgated by the supreme court).

(n85) Of the nine modified comparative fault states that were the original adopters of the Wisconsin model, see states listed supra note 76, only Ohio, McClure v. Neuman, 178 N.E.2d 621 (Ohio Ct. App. 1961); a watered down Texas, see supra note 84; and Wisconsin remain with the blindfold rule.

Two pure comparative fault states, Kentucky, Robinson v. Murlin Phillips & MFA Ins. Co., 557 S.W.2d 202 (Ky. 1977), and Michigan, Mitchell v. Perkins, 54 N.W. 293 (Mich. 1952), have taken positions against informing juries about the legal consequences of their findings. However, these states have no percentage bar, thus jurors there, who are prone to speculate on the law in comparative fault cases, are likely to guess right.

(n86) See supra note 76 and accompanying text.

(n87) 517 P.2d 416 (Colo. Ct. App. 1973), rev'd, 526 P.2d 298 (Colo. 1974).

(n88) Id. at 418.

(n89) Id.

(n90) Id.

(n91) Id.

(n92) Id.

(n93) Schaffer, supra note 7, at 837 n.68.

(n94) Simpson v. Anderson, 526 P.2d 298, 299 (Colo. 1974).

(n95) Thomas v. Wadlingten, 526 P.2d 295, 297 (Colo. 1974).

(n96) Id.

(n97) COLO. REV. STAT. Section 13-21-111(4) (1975) (currently codified at Id. Section 13-21-111.5(5) (1996)).

(n98) See infra notes 152-54 and accompanying text.

(n99) 1973 Wyo. Sess. Laws ch. 28 (codified at WYO. STAT. Section 1-7.2 (Michie Supp. 1973)). The statute was amended and renumbered in 1977, amended again in 1986, and again in 1994. Versions of the statute prior to and following the 1994 amendments are reproduced in Burman, supra note 2, in Appendixes A and B respectively.

(n100) Wyo. Stat. Section 7.2(a)was identical to the Wisconsin provision. Section 7.2(b), however, inadvertently omitted the requirement that, in a jury trial, the judge is to reduce the amount of damages in proportion to the plaintiffs negligence. See Smith, supra note 7, at 200 n.5 (quoting and discussing WYO. STAT. Section 1-7.2(b)).

(n101) 564 P.2d 844, 846 (Wyo. 1977).

(n102) "We hold that the intent of the Wyoming State Legislature was to adopt the Wisconsin judicial construction of the comparative negligence statute at the date of enactment, holding that is reversible error to advise the jury in argument or by instruction of the effect of its verdict." Id.

(n103) Smith, supra note 7, at 21.6.

(n104) Id.

(n105) WYO. STAT. Section 1-7.7 (1976 Interim Supp.) (cited in Johnson v. Safeway Stores, Inc., 568 P.2d 908, 911 (Wyo. 1977) (emphasis added).

(n106) 568 P.2d 908 (Wyo. 1977).

(n107) Id. at 913.

(n108) Id. at 912-13. The court observed that, when Woodward was decided, Section 1-7.7 had been enacted, but after the verdict had been rendered, the express intent of the legislature was to apply the statute prospectively.

(n109) Id. at 913-14.

(n110) Id. at 914.

(n111) WYO. STAT. ANN. Section 1-1-109(b) (Michie 1988).

(n112) Wyo. Stat. Ann. Section 1-1-109(c)(i)(B) (Michie 1996).

(n113) See supra notes 8-12 and accompanying text.

(n114) Burman, supra note 2, at 552.

(n115) Id. at 531.

(n116) For accounts of the Illinois experience with comparative fault principles through the 1970s, see Alvis v. Ribar, 421 N.E.2d 886 (Ill. 1981); Carol Isackson, Note and Comment, Pure Comparative Negligence in Illinois, 58 CHI.-KENT L. REV. 599 (1982).

(n117) See infra notes 122-28 and accompanying text.

(n118) See infra notes 139-42 and accompanying text.

(n119) See infra notes 143-45 and accompanying text.

(n120) See infra notes 146-47 and accompanying text.

(n121) See infra notes 148-57 and accompanying text.

(n122) 20 Ill. 478 (1858).

(n123) Id. at 497.

(n124) Isackson, supra note 116, at 604 n.28.

(n125) 421 N.E.2d at 889 (citing Chicago, Milwaukee & St. Louis Ry. Co. v. Mason, 27 Ill. pp. 450, 453, 454 (1888)); Isackson, supra note 116, at 603.

(n126) 421 N.E.2d at 889; Isackson, supra note 116, at 604.

(n127) 38 N.E. 892 (Ill. 1894).

(n128) Id. at 893.

(n129) 421 N.E.2d at 890 (discussing the defense's unavailability to a "willful, wanton, or reckless" defendant, to a defendant who violated a statute enacted "to protect a plaintiff from his own improvident acts," and to one who had the "last clear chance" to avoid the injury; and although the latter doctrine was repudiated by Illinois courts it was employed nevertheless without labeling).

(n130) Isackson, supra note 1.16, at 605.

(n131) See supra notes 39-50 and accompanying text.

(n132) See Maki v. Frelk, 239 N.E.2d 445 (Ill. 1968) (Ward, J., dissenting). The plaintiff-appellant declares, without contradiction being offered, that since 1937 there have been nine attempts in our legislature to change our contributory negligence rule and that with a single exception none reached the floor of either House. The prospect of securing through legislation a rule better styled to achieve fair dispositions in negligence cases does not appear to be bright. Id. at 450.

(n133) See SCHWARTZ, supra note 6, Section 1.4(B), at 13.

(n134) Id. at 13-14.

(n135) Id. Subsection 1.5(D) & (E), at 21-27.

(n136) See cases cited for this proposition from N.M., Mich., W. Va., Cal., Alaska and Fla. in Alvis v. Ribar, 421 N.E.2d 886, 895 (Ill. 1981).

(n137) See supra note 55 and accompanying text.

(n138) Nga Li v. Yellow Cab Co., 532 P.2d 1226 (Cal. 1975).

(n139) 229 N.E.2d 284 (Ill. App. Ct. 1967), rev'd, 239 N.E.2d 445 (Ill. 1968).

(n140) Id. at 290.

(n141) See Isackson, supra note 116, at 609 (noting that "in a Vanderbilt University Symposium, five out of the six participants favored the idea of judicial change"); Comment, Comments on Maki v. Frelk -- Comparative v. Contributory Negligence: Should the Court or Legislature Decide?, 21 VAND. L. REV. 889, 899 (1968).

(n142) 239 N.E.2d at 447.

(n143) 421 N.E.2d 886 (Ill. 1981).

(n144) Id. at 898. The court disagreed with the defense's contention that the failure of the Illinois General Assembly to enact any of the six bills introduced since 1976 was "a sign of the General Assembly's desire to retain the present [contributory negligence] status of the rule." Id. at 895. Rather, the inaction can be attributed "`to its feeling that it is more appropriate, considering the history of the question in Illinois, for the judiciary to act.'" Id. (quoting Maki, 239 N.E.2d at 450 (Ward, J., dissenting)).

We believe that the proper relationship between the legislature and the court is one of cooperation and assistance in examining and changing the common law to conform with the

ever-changing demands of the community. There are, however, times when there exists a mutual state of inaction in which the court awaits action by the legislature and the legislature awaits guidance from the court. Such a stalemate is a manifest injustice to the public.

Id. at 896.

(n145) Id. at 898.

(n146) 735 ILL. COMP. STAT. 5/2-1107,1, 5/2-1116, 5/2-1117 (West 1990) (as renumbered).

(n147) Id. at Section 5/2-1107.1.

(n148) See, e.g., David Frum, Tort Reform, One Step Forward, One Step Back, WALL ST. J., May 27, 1992, at A15; Ellen J. Pollack, Quayle Calls for Changes in Legal System, WALL ST. J., Aug. 14, 1991, at B10; Aric Press et al., Are Lawyers Burning America?, NEWSWEEK, Mar. 20, 1995, at 32.

(n149) Illinois Civil Justice Reform Act of 1995, Pub. Act No. 89-7, 1995 Ill. Legis. Serv. 224 (West) (codified in scattered sections of ILL. COMP. STAT. chs. 430, 730, 735, 740, 745, 815, 820 (West Supp. 1996). The Act is reprinted with strikeouts from, and additions to, the old statute at 27 LOY. U. CHI. L.J. 819 (1996). See Martha Middleton, A Changing Landscape as Congress Struggles to Rewrite the Nation's Tort Laws, The States Already may Have Done the Job, A.B.A. J., Aug. 1995, at 57 ("The new Civil Justice Reform Amendments of 1995 are touted as the most comprehensive changes in tort law adopted by a state legislature."). For overviews of the Act, see Michael Gallegher et al., Illinois Tort Reform: The Judge's Perspective, 84 ILL B.J. 124 (1996); J. Jeffrey Zimmerman et al., A Review of the Illinois Civil Justice Reform Act of 1995, 83 ILL. B.J. 282 (1995).

(n150) See Kirk W. Dillard, Illinois' Landmark Tort Reform: The Sponsor's Policy Explanation, 27 LOY. U. CHI. L.J. 803, 814-16 (1996).

(n151) Id. at 813-14.

(n152) 735 ILL. COMP. STAT. 5/21107.1 (West 1990) (as renumbered).

(n153) The old "sunshine" section 5/2-1107.1, id., with strikeouts and additions to convert it into the new "blindfold" section is reproduced in its entirety supra note 12 (emphasis added).

(n154) See supra note 85 and accompanying text.

(n155) 735 ILL. COMP. STAT. 5/2-1107.1 (West 1996) (emphasis added).

(n156) Gallegher et al., supra note 149, at 130.

(n157) Id.

(n158) The empirical work that has been done on jury decision-making is concentrated in the social science literature. For two surveys of this literature, its contributions and limitations, see Nancy Pennington & Reid Hastie, Practical Implications of Psychological Research on Juror and Jury Decision Making, 16 PERSONALITY & SOC. PSYCHOL. BULL. 90 (1990); Nancy Pennington & Reid Hastie, Jury Decision-Making Models: The Generaliza-tion Gap, 89 PSYCHOL. BULL. 246 (1981). See also REID HASTIE ET AL., INSIDE THE JURY (1983); Franklin J. Boster et al., An Information-processing Model of Jury Decision Making; 18 COMM. RES. 524 (1991). Most of this work has focused on the criminal jury process: see, e.g., Mark Constanzo & Sally Constanzo, Jury Decision Making in the Capital Penalty Phase, 16 LAW & HUM. BEHAV. 185 (1992); Richard L. Wiener et al., Comprehensibility of Approved Jury Instructions in Capital Murder Cases, 80 J. APPLIED PSYCHOL. 455 (1995). Much of the empirical work published to date studies the effects of personal attributes of jurors, race and socio-economic background, or deliberation style on jury decisions: see. e.g., Jeffrey R. Boyll, Psychological, Cognitive, Personality and Interpersonal Factors in Jury Verdicts, 15 LAW & PSYCHOL. REV. 163 (1991); Jane Goodman et al., Money, Sex, and Death: Gender Bias in Wrongful Death Awards, 25 LAW & SOC'Y REV. 263 (1991). Relatively few works consider the effects of differences in legal rules on jury decisions, but see Irwin A. Horowitz, Jury Nullification: The Impact of Judicial Instructions, Arguments and Challenges on Jury Decision Making, 12 LAW & HUM. BEHAV. 439 (1988); Ronald J. Matlon et al., Factors Affecting Jury Decision-making, 12 SOC. ACTION & L. 41 (1986):

It is somewhat surprising that relatively little empirical research has been devoted to comparing the impact of extralegal factors to the various essential content-oriented (legal) aspects of the trial itself on the decisions of the juries. Essential content-oriented aspects of a trial may be described as the established conventions of the court which constitute the substance of the trial, such as opening statements, direct testimony by witnesses, direct and cross-examination from counsel, objections, exhibits, closing arguments, and judge's instructions.

Id. at 41-42.

(n159) The transcripts of the two simulation versions of McKey v. Torino Pizza Inc., including instructions and verdict forms, and exhibits are on file with the authors at the office of Professor Jordan H. Leibman, Kelley School of Business, Indiana University, 801 W. Michigan St., Indianapolis IN, 46202.

(n160) We would have preferred uniform juries of six persons, the standard civil jury size in Indiana, but our viewing groups of subjects rarely proved to be evenly divisible by six.

(n161) The Base juries rendered what were essentially "special verdicts" because, after determining percentages of fault and amount of damages by the juries, it would be necessary for the court to adjust these verdicts to conform to the 51% percentage bar. The Variant juries rendered "general verdicts with special interrogatories" in which the jury determines the ultimate outcome of their findings. See supra notes 66-69 and accompanying text for an explanation of these two forms of verdict.

(n162) In exchange for the use of the physical evidence in possession of the plaintiffs attorney, we agreed to protect the privacy of the parties by withholding all names, the exact locations of the relevant incidents, and the name of the case. In this article, we refer to the actual case upon which the simulation was based as the "underlying case."

(n163) These facts and those in the following paragraphs are essentially the same in the underlying case.

(n164) In the underlying case there was virtually no possibility for the plaintiff to recover damages from the State. IND. CODE Sections 34-4-33-8 (Supp. 1984) provides that the Comparative Fault Act "does not apply . . . to tort claims against governmental entities .... " With respect to the State, contributory negligence remains a complete defense. Under the facts of McKey, some negligence on Dawn's part is conceded by the plaintiff. Presumably, the State was named as a party defendant to attract a portion of the fault that might otherwise be attributed to Dawn. Because Indiana is a modified comparative fault state with a 51% bar to recovery, see supra note 6, it was important for the plaintiff to make it as easy as possible for the jury to assign at least 50% to the defendants. The defense was probably ambivalent about this strategy because it could not be confident that Dawn would be found more than 50% responsible for her own injuries. It, therefore, may have welcomed the opportunity to lay off part of the fault on another defendant In any event, the jury knew none of this and had to assume that the State would pay its share of any damages assessed against it.

(n165) See infra notes 183-85 and accompanying text.

(n166) Standard deviation was 6.3.

(n167) Only one juror was 17, the remainder were "jury-eligible" with respect to age. About 11% were at least 30 years of age; 65% were between 19 and 22 years old. These student juries were no doubt younger than those typically encountered in the "real world," but they were older on average than might be surmised from their sophomore/junior status.

(n168) Simulated Trial Transcript, McKey v. Torino Pizza Co., Inc., Variant version at INSTRUC7-1 (on file with the authors, see supra note 159).

(n169) Id. at CLOS7-6.

(n170) Id. at CLOS7-21.

(n171) Two of the Variant juries were hung. Their verdict data is not included in the analysis.

(n172) As it turned out, Dawn's negligence was deemed by the average blindfold jury to be significantly higher than 50%. See Table 3. The mean percentage of fault attributable to her by these juries -- who were unaware that a 51% or higher finding would bar her -- was 65.6%. This relatively high degree of negligence proved to be a significant hurdle for those Variant juries that wished to award her damages. See infra note 191 and accompanying text for a discussion of this problem.

(n173) See supra text accompanying and surrounding notes 166-67 (discussing the origin and demographics of the subjects who served as mock jurors).

(n174) See supra note 6 and accompanying text.

(n175) Id.

(n176) See infra Table 1, col. 10.

(n177) See infra Table 3, col. 6.

(n178) This test (a.k.a. 2-sample difference of means test) is appropriate when one wishes to compare two observed sample means to statistically assess whether two different samples came from populations having equal means. DONALD R. HARNETT, STATISTICAL METHODS 376 (3d ed.1982). In this instance we used the t-distribution and the corresponding t-test difference of means, because we operate under the assumption that the population variances are not known. Inasmuch as we used the sample variances, the t-test is appropriate. For a more thorough discussion of this test, see HARNETT, supra, at 345-80; UMA SEKARAN, RESEARCH METHODS FOR BUSINESS: A SKILL BUILDING APPROACH 249-50 (2d ed. 1992).

(n179) See infra Table 1, col. 6.

(n180) See infra Table 3, col. 6.

(n181) See supra Table 1, col. 8.

(n182) See supra Table 3, col. 8.

(n183) See supra note 165 and accompanying text. The plaintiff sought to have fault allocated to the State sufficient to produce an award, but not so much as to excessively dissipate collectible damages (the State being essentially immune from having to pay damages, see supra note 164 and accompanying text). The defense was interested in shifting its fault to the State, but not so much as to bring Dawn's fault below 51%.

(n184) See infra Table 4, col. 12.

(n185) See infra Table 5, col. 12.

(n186) Of the sunshine jurors, 12.6% neither agreed nor disagreed; 49.8% disagreed or strongly disagreed with the statement. We were surprised how many jurors felt this influence, were prepared to act on it, and were willing to admit that they were acting beyond the judicial charge.

(n187) Of these jurors, 15.4% neither agreed nor disagreed, while 47.3% disagreed or strongly disagreed.

(n188) See supra Table 2; 12 of 39 blindfold juries, or 31%, produced an award for Dawn.

(n189) See supra Table 3; 17 of 43 sunshine juries, or 40%, produced an award.

(n190) The z-test difference of proportions (a.k.a. 2-sample difference of proportions test) is appropriate when one wishes to compare two observed sample proportions to statistically assess whether two different samples came from populations having equal proportions. HARNETT, supra note 178, at 401. For a more thorough discussion of this test, see id. at 401-02; WILLIAM R. DILLON ET AL., ESSENTIALS OF MARKETING RESEARCH 418-20 (1993).

(n191) See supra Table 1, col. 12.

(n192) Running more iterations for both samples might settle this question.

(n193) See supra Table 3, col. 11. The mean award to Dawn is the sum of the damages assessed against Torino and the State of Indiana.

(n194) See supra Table 2, col. 1.1.

(n195) See infra Table 6, col. 11.

(n196) See supra Table 7, col. 11.

(n197) The Median as a Measure of Central Tendency. Although policy makers are more likely to be concerned about mean results, practitioners are likely to be more interested in the typicality of an experimental outcome. If, for example, an attorney is representing a client like Torino Pizza Company, he will want to know what the typical verdict result in such cases has been. With respect to the percentage of fault attributed to Torino by the blindfold sample, the data are positively skewed. Torino's mean fault in the blindfold sample is 16.3%, but the median is 12%. The attorney should probably tell his client to contemplate a 12% finding, rather than 16.3%, but he should caution the client that: some large findings of fault by blindfold juries against corporate defendants have been recorded. For each of the parameters we have so far considered, both of these measures of central tendency should be considered.

(n198) See PROSSER AND KEETON, supra note 23, Section 67, at 469 ("Courts have become more reluctant to rule that the plaintiff's conduct is negligent as a matter of law, and juries are notoriously inclined to find there has been no such negligence, or to make some more or less haphazard reduction of the plaintiffs damages in proportion to his fault.").

(n199) See supra Table 1, col. 11.

(n200) See supra Table 2, col. 11.

(n201) See supra Table 3, col. 11.

(n202) See supra notes 194-96 and accompanying text.

(n203) Daniel R. Ilgen, Laboratory Research: A Question of When, Not If, in GENERALIZING FROM LABORATORY TO FIELD SETTINGS 257, 257-58 (Edwin E. Locke ed. 1986). For further commentary on the lab versus field issue, see John P. Cambell, Labs, Fields, and Straw Issues and Generalizing from Laboratory to Field: Ecological Validity or Abstraction of Essential Elements, in GENERALIZING FROM LABORATORY TO FIELD SETTINGS, supra, at 269 and 3 respectively.

(n204) Id.

(n205) Id. at 258.

(n206) See supra text accompanying note 162.

(n207) On a scale of 1 to 5 (1 = strongly agree) the mean responses to the 5 items were 2.44, 2.49, 1.97, 1.90, 1.88 respectively. Jurors also reported that they had plenty of time to reach a verdict. Exit survey forms are on file with the authors, see supra note 159.

(n208) See supra text accompanying note 203,

(n209) See GENERALIZING FROM LABORATORY TO FIELD SETTINGS (Edwin A. Locke ed. 1986) (citing numerous instances from the behavioral literature where lab studies have been replicated in the field with similar results).

(n210) For an outstanding example of the risk-taking mentality of some tort lawyers, see HARR, supra note 69.

(n211) The authors have plans to replicate this experiment using subjects recruited from the jury rolls of Hendricks County, Indiana.

(n212) Best v. Taylor Mach. Works, Nos. 81890, 81891, 81892, 81893 cons., 1997 Ill. LEXIS 478, at *155 (IlL Dec. 18, 1997).

(n213) "The cap on compensatory damages for noneconomic injury is, as the parties acknowledge, at the heart of [Civil Justice Reform Act] Public Act 89-7." Id. at *17 (discussing 735 ILL. COMP. STAT. 5/2-1115.1(a) (West 1996)).

(n214) Id. at *155 (discussing 735 ILL COMP. STAT. 5/2-1107.1) (West 1997).

(n215) Id. at *156.

(n216) Id. at *155.

(n217) Shortly after the supreme court struck down the Civil Justice Reform Act, a spokesman for the Speaker of Illinois House of Representatives was reported to have said "that the speaker would be willing to discuss legislation that would re-enact provisions the Supreme Court had not singled out and ruled invalid." Supporters of the Act were reported to have "voiced skepticism

regarding their chances if the Chicago Democrat remains in control of the Illinois House of Representatives." David Heckleman, As Dem Leader Ponders Reviving Some Reforms, CHICAGO DAILY L. BULL, Dec. 19, 1997, available in LEXIS, News Library, Curnws File.

(n218) Best, 1997 Ill. LEXIS 478, at *17-23 & n.1 (discussing affidavits submitted by Professor Nell Vidmar of the Duke Law School, Professor Marc Galanter, of the University of Wisconsin Law School, and a joint affidavit from Stephen Daniels, a senior research fellow at the American Bar Foundation and Joanne Martin, an assistant director of the Foundation).

(n219) Professor Vidmar cited an Indiana study that showed an increase in malpractice claims following the capping of malpractice damages, Id. at *20 n.1; Professor Galanter submitted data showing that court filings in Cook County actually declined during the period 1980-94 (prior to the tort reform legislation) Id. at *20; and Daniels and Martin summarized empirical studies showing "that only a tiny fraction of accidental deaths and injuries are pursued through the litigation system as claims for compensation," Id. at *22.

(n220) Id. at *23.

(n221) Id. at *18.

(n222) Id. at *23-24.

TABLE 1 MCKEY V. TORINO PIZZA CO., INC. BASE ("BLINDFOLD") SAMPLE DATA BEFORE ADJUSTMENT FOR MOD. COMP. FAULT 1 2 3 4 5 6 $7 NO. PANEL DATE VERS HU TORINO TOR-DAMAGES % $ 1 B101 7/17 B 12.0% $240,000 2 B109 10/12 B 34.0% $899,314 3 B110 10/12 B 0.0% $0 4 B111 10/12 B 40.0% $1,400,000 5 B112 10/12 B 0.0% $0 6 B113 10/12 B 35.0% $2,100,000 7 B114 10/12 B 5.0% $92,500 8 C101 9/7 B 30.0% $802,050 9 C102 9/7 B 5.0% $272,200 10 C103 9/7 B 20.0% $450,000 11 C104 9/7 B 23.0% $1,369,644 12 C105 9/7 B 10.0% $336,960 13 C106 9/7 B 30.0% $1,357,572 14 E101 9/10 B 15.0% $267,974 15 E102 9/10 B 15.0% $828,786 16 E103 9/10 B 15.0% $375,000 17 E104 9/10 B 0.0% $0 18 E105 9/10 B 30.0% $1,830,000 19 E106 9/10 B 0.0% $0 20 F101 9/20 B 0.0% $0 21 F102 9/20 B 5.0% $11,005 22 G201 10/15 B 20.0% $600,000

23 G202 10/5 B 10.0% $600,000 24 H101 10/20 B 30.0% $1,290,000 25 H102 10/20 B 50.0% $2,993,760 26 H103 10/20 B 55.0% $3,300,000 27 H104 10/20 B 0.0% $0 28 H105 10/20 B 5.0% $360,000 29 H106 10/20 B 10.0% $22,010 30 H107 10/20 B 15.0% $900,000 31 H108 10/20 B 15.0% $155,087 32 H109 10/20 B 0.0% $0 33 H110 10/20 B 0.0% $0 34 H111 10/20 B 0.0% $0 35 H112 10/20 B 10.0% $280,354 36 H113 10/20 B 10.0% $100,000 37 H114 10/20 B 0.0% $0 38 H115 10/20 B 25.0% $717,400 39 H116 10/20 B 55.0% $3,685,000 MEANS 16.3% $708,631 MEDIANS 12.0% $336,960 ST. DEV. 0.159 $946,528 1 8 9 10 11 NO. STATE ST-DAMAGES TOR+ST TOR+ST % $ % $ 1 8.0% $160,000 20.0% $400,000 2 0.0% $0 34.0% $899,314 3 66.0% $2,970,000 66.0% $2,970,000 4 20.0% $700,000 60.0% $2,100,000 5 5.0% $173,643 5.0% $173,643 6 0.0% $0 35.0% $2,100,000 7 35.0% $647,500 40.0% $740,000 8 30.0% $802,050 60.0% $1,604,100 9 5.0% $272,200 10.0% $544,400 10 50.0% $1,125,000 70.0% $1,575,000 11 23.0% $1,369,644 46.0% $2,739,287 12 20.0% $673,920 30.0% $1,010,880 13 40.0% $1,810,096 70.0% $3,167,668 14 15.0% $267,974 30.0% $535,948 15 35.0% $1,933,834 50.0% $2,762,620 16 10.0% $250,000 25.0% $625,000 17 0.0% $0 0.0% $0 18 40.0% $2,440,000 70.0% $4,270,000 19 10.0% $122,010 10.0% $122,010 20 10.0% $174,485 10.0% $174,485 21 10.0% $22,010 15.0% $33,015 22 5.0% $150,000 25.0% $750,000 23 40.0% $2,400,000 50.0% $3,000,000 24 10.0% $430,000 40.0% $1,720,000 25 15.0% $898,128 65.0% $3,891,888 26 5.0% $300,000 60.0% $3,600,000 27 0.0% $0 0.0% $0 28 5.0% $360,000 10.0% $720,000 29 20.0% $44,020 30.0% $66,030 30 35.0% $2,100,000 50.0% $3,000,000

31 15.0% $155,087 30.0% $310,173 32 10.0% $452,524 10.0% $452,524 33 10.0% $344,637 10.0% $344,637 34 0.0% $0 0.0% $0 35 25.0% $700,885 35.0% $981,239 36 25.0% $250,000 35.0% $350,000 37 20.0% $689,274 20.0% $689,274 38 0.0% $0 25.0% $717,400 39 35.0% $2,345,000 90.0% $6,030,000 MEANS 18.1% $705,998 34.4% $1,414,629 MEDIANS 15.0% $344,637 30.0% $740,000 ST. DEV. 0.159 $828,617 23.4% $1,456,807 1 12 13 14 NO. DAWN DAWN-DAMAGES TOTAL DAMAGES % $ $ 1 80.0% $1,600,000 $2,000,000 2 66.0% $1,745,726 $2,645,040 3 34.0% $1,530,000 $4,500,000 4 40.0% $1,400,000 $3,500,000 5 95.0% $3,299,217 $3,472,860 6 65.0% $3,900,000 $6,000,000 7 60.0% $110,000 $1,850,000 8 40.0% 9 90.0% $41,899,600 $5,444,000 10 30.0% $675,000 $2,250,000 11 54.0% $3,215,685 $5,954,972 12 70.0% $2,358,720 $3,369,600 13 30.0% $1,357,572 $4,525,240 14 70.0% $1,250,544 $1,786,492 15 50.0% $2,762,620 $5,525,240 16 75.0% $1,875,000 $2,500,000 17 100.0% $0 $0 18 30.0% $1,830,000 $6,100,000 19 90.0% $1,098,090 $1,220,100 20 90.0% $1,570,365 $1,774,850 21 85.0% $187,085 $220,100 22 75.0% $2,250,000 $3,000,000 23 50.0% $3,000,000 $6,000,000 24 60.0% $2,580,000 $4,300,000 25 35.0% $2,095,632 $5,987,520 26 40.0% $2,400,000 $6,000,000 27 100.0% $0 $0 28 90.0% $6,480,000 $7,200,000 29 70.0% $154,070 $220,100 30 50.0% $3,000,000 $6,000,000 31 70.0% $723,738 $1,033,911 32 90.0% $4,072,716 $4,525,240 33 90.0% $3,101,733 $3,446,370 34 100.0% $0 $0 35 65.0% $1,822,302 $2,803,541 36 65.0% $650,000 $1,000,000 37 80.0% $2,757,096 $3,446,370 38 75.0% $2,152,200 $2,869,600

39 10.0% $670,000 $6,700,000 MEANS 65.6% $1,965,234 $3,379,863 MEDIANS 70.0% $1,822,302 $3,369,600 ST. DEV. 23.4% $1,402,757 $2,120,585 TABLE 2 MCKEY V. TORINO PIZZA CO., INC. BASE ("BLINDFOLD") SAMPLE DATA AFTER ADJUSTMENT FOR MOD. COMP. FAULT 1 2 3 4 5 6 7 NO. PANEL DATE VERS. HU TORINO TOR-DAMAGES % $ 1 B101 7/17 B 12.0% $0 2 B109 10/12 B 34.0% $0 3 B110 10/12 B 0.0% $0 4 B111 10/12 B 40.0% $1,400,000 5 B112 10/12 B 0.0% $0 6 B113 10/12 B 35.0% $0 7 B114 10/12 B 5.0% $0 8 C101 9/7 B 30.0% $802,050 9 C102 9/7 B 5.0% $0 10 C103 9/7 B 20.0% $450,000 11 C104 9/7 B 23.0% $0 12 C105 9/7 B 10.0% $0 13 C106 9/7 B 30.0% $1,357,572 14 E101 9/10 B 15.0% $0 15 E102 9/10 B 15.0% $828,786 16 E103 9/10 B 15.0% $0 17 E104 9/10 B 0.0% $0 18 E105 9/10 B 30.0% $1,830,000 19 E106 9/10 B 0.0% $0 20 F101 9/20 B 0.0% $0 21 F102 9/20 B 5.0% $0 22 G201 10/5 B 20.0% $0 23 G202 10/5 B 10.0% $600,000 24 H101 10/20 B 30.0% $0 25 H102 10/20 B 50.0% $2,993,760 26 H103 10/20 B 55.0% $3,300,000 27 H104 10/20 B 0.0% $0 28 H105 10/20 B 5.0% $0 29 H106 10/20 B 10.0% $0 30 H107 10/20 B 15.0% $900,000 31 H108 10/20 B 15.0% $0 32 H109 10/20 B 0.0% $0 33 H110 10/20 B 0.0% $0 34 H111 10/20 B 0.0% $0 35 H112 10/20 B 10.0% $0 36 H113 10/20 B 10.0% $0 37 H114 10/20 B 0.0% $0 38 H115 10/20 B 25.0% $0 39 H116 10/20 B 55.0% $3,685,000 MEANS 16.3% $465,312 MEDIANS 12.0% $0 ST. DEV. 0.159 $957,550

1 8 9 10 11 NO. STATE ST-DAMAGES TOR+ST TOR+ST % $ % $ 1 8.0% $0 20.0% $0 2 0.0% $0 34.0% $0 3 66.0% $2,970,000 66.0% $2,970,000 4 20.0% $700,000 60.0% $2,970,000 5 5.0% $0 5.0% $0 6 0.0% $0 35.0% $0 7 35.0% $0 40.0% $0 8 30.0% $802,050 60.0% $1,604,100 9 5.0% $0 10.0% $0 10 50.0% $1,125,000 70.0% $1,575,000 11 23.0% $0 46.0% $0 12 20.0% $0 30.0% $0 13 40.0% $1,810,096 70.0% $3,167,688 14 15.0% $0 30.0% $0 15 35.0% $1,933,834 50.0% $2,762,620 16 10.0% $0 25.0% $0 17 0.0% $0 0.0% $0 18 40.0% $2,440,000 70.0% $4,270,000 19 10.0% $0 10.0% $0 20 10.0% $0 10.0% $0 21 10.0% $0 15.0% $0 22 5.0% $0 25.0% $0 23 40.0% $2,400,000 50.0% $3,000,000 24 10.0% $0 40.0% $0 25 15.0% $898,128 65.0% $3,891,888 26 5.0% $300,000 60.0% $3,600,000 27 0.0% $0 0.0% $0 28 5.0% $0 10.0% $0 29 20.0% $0 30.0% $0 30 35.0% $2,100,000 50.0% $3,000,000 31 15.0% $0 30.0% $0 32 10.0% $0 10.0% $0 33 10.0% $0 10.0% $0 34 0.0% $0 0.0% $0 35 25.0% $0 35.0% $0 36 25.0% $0 35.0% $0 37 20.0% $0 20.0% $0 38 0.0% $0 25.0% $0 39 35.0% $2,345,000 90.0% $6,030,000 MEANS 18.3% $508,310 34.4% $973,622 MEDIANS 15.0% $0 30.0% $0 ST. DEV. 0.157 $898,514 0.234 $1,619,631 1 12 13 14 NO. DAWN DAWN-DAMAGES TOTAL DAMAGES % $ $ 1 80.0% $0 $0 2 66.0% $0 $0 3 34.0% $1,530,000 $4,500,000 4 40.0% $1,400,000 $3,500,000

5 95.0% $0 $0 6 65.0% $0 $0 7 60.0% $0 $0 8 40.0% $1,069,400 $2,673,500 9 90.0% $0 $0 10 30.0% $675,000 $2,250,000 11 54.0% $0 $0 12 70.0% $0 $0 13 30.0% $1,357,572 $4,525,240 14 70.0% $0 $0 15 50.0% $2,762,620 $5,525,240 16 75.0% $0 $0 17 100.0% $0 $0 18 30.0% $1,830,000 $6,100,000 19 90.0% $0 $0 20 90.0% $0 $0 21 85.0% $0 $0 22 75.0% $0 $0 23 50.0% $3,000,000 $6,000,000 24 60.0% $0 $0 25 35.0% $2,095,632 $5,987,520 26 40.0% $2,400,000 $6,000,000 27 100.0% $0 $0 28 90.0% $0 $0 29 70.0% $0 $0 30 50.0% $3,000,000 $6,000,000 31 70.0% $0 $0 32 90.0% $0 $0 33 90.0% $0 $0 34 100.0% $0 $0 35 65.0% $0 $0 36 65.0% $0 $0 37 80.0% $0 $0 38 75.0% $0 $0 39 10.0% $670,000 $6,700,000 MEANS 65.6% $558,724 $1,532,346 MEDIANS 70.0% $0 $0 ST. DEV. 0.234 $961,769 $2,460,698 TABLE 3 MCKEY V. TORINO PIZZA CO., INC. VARIANT ("SUNSHINE") SAMPLE DATA 1 2 3 4 5 6 $7 NO. PANEL DATE VERS. HU TORINO TOR-DAMAGES % $ 1 B103 10/5 V3 60.0% $6,000,000 2 B104 10/5 V3 10.0% $0 3 B105 10/5 V3 40.0% $0 4 B106 10/5 V3 25.0% $0 5 B107 10/5 V3 15.0% $0 6 B108 10/5 V3 49.0% $2,524,625 7 C107 9/6 V3 40.0% $2,400,000 8 C108 9/6 V3 HU 9 C109 9/6 V3 30.0% $0 10 C110 916 V3 44.0% $2,640,000

11 C111 9/6 V3 40.0% $1,400,000 12 C112 9/6 V3 20.0% $0 13 C113 9/6 V3 25.0% $0 14 D101 8/14 V3 35.0% $0 15 D102 8/14 V3 10.0% $0 16 D103 8/14 V3 HU 17 E201 9/11 V3 30.0% $0 18 E202 9/11 V3 10.0% $0 19 E201 9/11 V3 51.0% $892,316 20 E204 9/11 V3 40.0% $1,700,000 21 E205 9/11 V3 10.0% $172,010 22 E206 9/11 V3 50.0% $2,262,620 23 F201 9/20 V3 52.0% $1,458,937 24 F202 9/20 V3 11.0% $0 25 F203 9/20 V3 15.0% $0 26 F204 9/20 V3 48.0% $105,648 27 G101 10/5 V3 0.0% $0 28 G102 10/5 V3 15.0% $0 29 G103 10/5 V3 75.0% $5,519,861 30 G104 10/5 V3 30.0% $0 31 G105 10/5 V3 20.0% $117,647 32 I201 10/31 V3 30.0% $0 33 I202 10/31 V3 35.0% $0 34 I203 10/31 V3 9.0% $0 35 I204 10/31 V3 35.0% $0 36 I205 10/31 V3 0.0% $0 37 I206 10/31 V3 55.0% $3,850,000 38 I207 10/31 V3 25.0% $0 39 I208 10/31 V3 50.0% $2,977,486 40 I209 10/31 V3 23.0% $0 41 U101 1/23 V3 30.0% $0 42 U102 1/23 V3 25.0% $0 43 U103 1/23 V3 15.0% $300,000 44 U104 1/23 V3 20.0% $0 45 U105 1/23 V3 8.0% $64,000 MEANS 29.3% $799,655 MEDIANS 30.0% $0 ST. DEV. 17.4% $1,502,432 1 8 9 10 11 NO. STATE ST-DAMAGES TOR+ST TOR+ST % $ % $ 1 10.0% $1,000,000 70.0% $7,000,000 2 30.0% $0 40.0% $0 3 9.0% $0 49.0% $0 4 24.0% $0 49.0% $0 5 33.0% $0 48.0% $0 6 10.0% $515,230 59.0% $3,039,854 7 20.0% $1,200,000 60.0% $3,600,000 8 9 10.0% $0 40.0% $0 10 25.0% $1,500,000 69.0% $4,140,000 11 10.0% $350,000 50.0% $1,750,000 12 20.0% $0 40.0% $0

13 0.0% $0 25.0% $0 14 0.0% $0 35.0% $0 15 15.0% $0 25.0% $0 16 17 10.0% $0 40.0% $0 18 20.0% $0 30.0% $0 19 1.0% $17,496 52.0% $909,813 20 20.0% $850,000 60.0% $2,550,000 21 41.0% $705,241 51.0% $877,251 22 20.0% $905,048 70.0% $3,167,668 23 10.0% $280,565 62.0% $1,739,502 24 11.0% $0 22.0% $0 25 5.0% $0 20.0% $0 26 2.0% $4,402 50.0% $110,050 27 15.0% $0 15.0% $0 28 10.0% $0 25.0% $0 29 0.0% $0 75.0% $5,519,861 30 10.0% $0 40.0% $0 31 31.0% $182,353 51.0% $300,000 32 10.0% $0 40.0% $0 33 0.0% $0 35.0% $0 34 40.0% $0 49.0% $0 35 10.0% $0 45.0% $0 36 10.0% $0 10.0% $0 37 10.0% $700,000 65.0% $4,550,000 38 15.0% $0 40.0% $0 39 10.0% $595,497 60.0% $3,572,983 40 2.0% $0 25.0% $0 41 0.0% $0 40.0% $0 42 5.0% $0 30.0% $0 43 36.0% $720,000 51.0% $1,020,000 44 10.0% $0 30.0% $0 45 43.0% $344,000 51.0% $408,000 MEANS 14.7% $229,531 44.0% $1,029,186 MEDIANS 10.0% $0 40.0% $0 ST. DEV. 11.6% $391,735 15.7% $1,774,573 1 12 13 14 NO. DAWN DAWN-DAMAGES TOTAL DAMAGES % $ $ 1 30.0% $3,000,000 $10,000,000 2 60.0% $0 $0 3 51.0% $0 $0 4 51.0% $0 $0 5 52.0% $0 $0 6 41.0% $2,112,441 $5,152,295 7 40.0% $2,400,000 $6,000,000 8 9 60.0% $0 $0 10 31.0% $1,860,000 $6,000,000 11 50.0% $1,750,000 $3,500,000 12 60.0% $0 $0 13 75.0% $0 $0 14 65.0% $0 $0

15 75.0% $0 $0 16 17 60.0% $0 $0 18 70.0% $0 $0 19 48.0% $839,827 $1,749,640 20 40.0% $1,700,000 $4,250,000 21 49.0% $842,849 $1,720,100 22 30.0% $1,357,572 $4,525,240 23 38.0% $1,066,146 $2,805,648 24 78.0% $0 $0 25 80.0% $0 $0 26 50.0% $110,050 $220,100 27 85.0% $0 $0 28 75.0% $0 $0 29 25.0% $1,839,954 $7,359,815 30 60.0% $0 $0 31 49.0% $288,235 $5,882,351 32 60.0% $0 $0 33 65.0% $0 $0 34 51.0% $0 $0 35 55.0% $0 $0 36 90.0% $0 $0 37 35.0% $2,450,000 $7,000,000 38 60.0% $0 $0 39 40.0% $2,381,989 $5,954,972 40 75.0% $0 $0 41 60.0% $0 $0 42 70.0% $0 $0 43 49.0% $980,000 $2,000,000 44 70.0% $0 $0 45 49.0% $392,000 $800,000 MEANS 56.0% $590,025 $1,619,210 MEDIANS 55.0% $0 $0 ST. DEV. 15.8% $904,109 $2,646,435 TABLE 4 MCKEY V. TORINO PIZZA CO., INC. BASE ("BLINDFOLD") SAMPLE DATA JURIES PRODUCING NO AWARD FOR PLAINTIFF 1 2 3 4 5 6 $7 8 TOR.- No. PANEL DATE VERS. HU TORINO DAMAGES STATE % $ % 1 E101 9/10 B 15.0% $0 15.0% 2 H106 10/20 B 10.0% $0 20.0% 3 B113 10/12 B 35.0% $0 0.0% 4 H114 10/20 B 0.0% $0 20.0% 5 H113 10/20 B 10.0% $0 25.0% 6 E104 9/20 B 0.0% $0 0.0% 7 H112 10/20 B 10.0% $0 25.0% 8 B112 10/12 B 0.0% $0 5.0% 9 B109 10/12 B 34.0% $0 0.0% 10 C102 9/7 B 5.0% $0 5.0% 11 E106 9/10 B 0.0% $0 10.0% 12 F101 9/20 B 0.0% $0 10.0%

13 F102 9/20 B 5.0% $0 10.0% 14 C104 9/7 B 23.0% $0 23.0% 15 H108 10/20 B 15.0% $0 15.0% 16 H101 10/20 B 30.0% $0 10.0% 17 C105 9/7 B 10.0% $0 20.0% 18 H104 10/20 B 0.0% $0 0.0% 19 H115 10/20 B 25.0% $0 0.0% 20 H109 10/20 B 0.0% $0 10.0% 21 G201 10/5 B 20.0% $0 5.0% 22 B114 10/12 B 5.0% $0 35.0% 23 E103 9/10 B 15.0% $0 10.0% 24 H110 10/20 B 0.0% $0 10.0% 25 H111 10/20 B 0.0% $0 0.0% 26 H105 10/20 B 5.0% $0 5.0% 27 B101 7/17 B 12.0% $0 8.0% MEANS 10.5% $0 11.0% MEDIANS 10.0% $0 10.0% ST. DEV. 0.110 $0 0.090 1 9 10 11 12 13 14 ST- DAWN- TOTAL No. DAMAGES TOR+ST TOR+ST DAWN DAMAGES DAMAGES $ % $ % $ $ 1 $0 30.0% $0 70.0% $0 $0 2 $0 30.0% $0 70.0% $0 $0 3 $0 35.0% $0 65.0% $0 $0 4 $0 20.0% $0 80.0% $0 $0 5 $0 35.0% $0 65.0% $0 $0 6 $0 0.0% $0 100.0% $0 $0 7 $0 35.0% $0 65.0% $0 $0 8 $0 5.0% $0 95.0% $0 $0 9 $0 34.0% $0 66.0% $0 $0 10 $0 10.0% $0 90.0% $0 $0 11 $0 10.0% $0 90.0% $0 $0 12 $0 10.0% $0 90.0% $0 $0 13 $0 15.0% $0 85.0% $0 $0 14 $0 46.0% $0 54.0% $0 $0 15 $0 30.0% $0 70.0% $0 $0 16 $0 40.0% $0 60.0% $0 $0 17 $0 30.0% $0 70.0% $0 $0 18 $0 0.0% $0 100.0% $0 $0 19 $0 25.0% $0 75.0% $0 $0 20 $0 10.0% $0 90.0% $0 $0 21 $0 25.0% $0 75.0% $0 $0 22 $0 40.0% $0 60.0% $0 $0 23 $0 25.0% $0 75.0% $0 $0 24 $0 10.0% $0 90.0% $0 $0 25 $0 0.0% $0 100.0% $0 $0 26 $0 10.0% $0 90.0% $0 $0 27 $0 20.0% $0 80.0% $0 $0 MEANS $0 21.5% $0 78.5% $0 $0 MEDIANS $0 25.0% $0 75.0% $0 $0

ST. DEV. $0 0.136 $0 0.136 $0 $0 TABLE 5 MCKEY V. TORINO PIZZA CO., INC. VARIANT ("SUNSHINE") SAMPLE DATA JURIES PRODUCING NO AWARD FOR PLAINTIFF 1 2 3 4 5 6 $7 8 TOR.- No. PANEL DATE VERS. HU TORINO DAMAGES STATE % $ % 1 C108 9/6 V3 HU 2 D103 8/14 V3 HU 3 E202 9/11 V3 10.0% $0 20.0% 4 U101 1/23 V3 30.0% $0 10.0% 5 B106 10/5 V3 25.0% $0 24.0% 6 B107 10/5 V3 15.0% $0 33.0% 7 I202 1/23 V3 35.0% $0 0.0% 8 U104 9/6 V3 20.0% $0 10.0% 9 C109 10/5 V3 30.0% $0 10.0% 10 G104 10/5 V3 30.0% $0 10.0% 11 B105 1/23 V3 40.0% $0 9.0% 12 U102 10/31 V3 25.0% $0 5.0% 13 I209 9/6 V3 23.0% $0 2.0% 14 C113 10/31 V3 25.0% $0 0.0% 15 I207 9/6 V3 25.0% $0 15.0% 16 D102 8/14 V3 10.0% $0 15.0% 17 F202 9/20 V3 11.0% $0 11.0% 18 E201 9/11 V3 30.0% $0 10.0% 19 F203 9/20 V3 15.0% $0 5.0% 20 I201 10/31 V3 30.0% $0 10.0% 21 I203 10/31 V3 9.0% $0 40.0% 22 C112 9/6 V3 20.0% $0 20.0% 23 D101 8/14 V3 35.0% $0 0.0% 24 G101 10/5 V3 0.0% $0 15.0% 25 I204 10/31 V3 35.0% $0 10.0% 26 B104 10/5 V3 10.0% $0 30.0% 27 I205 10/31 V3 0.0% $0 10.0% 28 G102 10/5 V3 15.0% $0 10.0% MEANS 21.3% $0 12.8% MEDIANS 10.0% $0 10.0% ST. DEV. 0.110 $0 0.101 1 9 10 11 12 13 14 ST- DAWN- TOTAL No. DAMAGES TOR+ST TOR+ST DAWN DAMAGES DAMAGES $ % $ % $ $ 1 2 3 $0 30.0% $0 70.0% $0 $0 4 $0 40.0% $0 60.0% $0 $0 5 $0 49.0% $0 51.0% $0 $0 6 $0 48.0% $0 52.0% $0 $0

7 $0 35.0% $0 65.0% $0 $0 8 $0 30.0% $0 70.0% $0 $0 9 $0 40.0% $0 60.0% $0 $0 10 $0 40.0% $0 60.0% $0 $0 11 $0 49.0% $0 51.0% $0 $0 12 $0 30.0% $0 70.0% $0 $0 13 $0 25.0% $0 75.0% $0 $0 14 $0 25.0% $0 75.0% $0 $0 15 $0 40.0% $0 60.0% $0 $0 16 $0 25.0% $0 75.0% $0 $0 17 $0 22.0% $0 78.0% $0 $0 18 $0 40.0% $0 60.0% $0 $0 19 $0 20.0% $0 80.0% $0 $0 20 $0 40.0% $0 60.0% $0 $0 21 $0 49.0% $0 51.0% $0 $0 22 $0 40.0% $0 60.0% $0 $0 23 $0 35.0% $0 65.0% $0 $0 24 $0 15.0% $0 85.0% $0 $0 25 $0 45.0% $0 55.0% $0 $0 26 $0 40.0% $0 60.0% $0 $0 27 $0 10.0% $0 90.0% $0 $0 28 $0 25.0% $0 75.0% $0 $0 MEANS $0 34.1% $0 65.9% $0 $0 MEDIANS $0 40.0% $0 62.5% $0 $0 ST. DEV. $0 0.110 $0 0.101 $0 $0 TABLE 6 MCKEY V. TORINO PIZZA CO., INC. VARIANT ("BLINDFOLD") SAMPLE DATA JURIES PRODUCING NO AWARD FOR PLAINTIFF 1 2 3 4 5 6 No. PANEL DATE VERS. HU TORINO % 1 H116 10/20 B 55.0% 2 C103 9/7 B 20.0% 3 C106 9/7 B 30.0% 4 E105 9/10 B 30.0% 5 B110 10/12 B 0.0% 6 H102 10/20 B 50.0% 7 C101 9/7 B 30.0% 8 B111 10/12 B 40.0% 9 H103 10/20 B 55.0% 10 H107 10/20 B 15.0% 11 E102 9/10 B 15.0% 12 G202 10/5 B 10.0% MEANS 29.2% MEDIANS 30.0% ST. DEV. 0.181

1 $7 8 9 10 No. TOR.-DAMAGES STATE ST-DAMAGES TOR+ST $ % $ % 1 $3,685,000 35.0% $2,345,000 90.0% 2 $450,000 50.0% $1,125,000 70.0% 3 $1,357,572 40.0% $1,810,096 70.0% 4 $1,830,000 40.0% $2,440,000 70.0% 5 $0 66.0% $2,970,000 66.0% 6 $2,993,760 15.0% $898,128 65.0% 7 $802,050 30.0% $802,050 60.0% 8 $1,400,000 20.0% $700,000 60.0% 9 $3,300,000 5.0% $300,000 60.0% 10 $900,000 35.0% $2,100,000 50.0% 11 $828,786 35.0% $1,933,834 50.0% 12 $600,000 40.0% $2,400,000 50.0% MEANS $1,512,264 34.3% $1,652,009 63.4% MEDIANS $1,128,786 35.0% $1,871,965 62.5% ST. DEV. $120,017 0.16 $625 0.114 1 11 12 13 14 No. TOR.-DAMAGES STATE ST-DAMAGES TOR+ST $ % $ % 1 $6,030,000 10.0% $670,000 $6,700,000 2 $15,750,001 30.0% $675,000 $2,250,000 3 $3,167,668 30.0% $1,357,572 $4,525,240 4 $4,270,000 30.0% $1,830,000 $6,100,000 5 $2,970,000 34.0% $1,530,000 $4,500,000 6 $3,891,888 35.0% $2,095,632 $5,987,520 7 $1,604,100 40.0% $1,069,400 $2,673,500 8 $2,100,000 40.0% $1,400,000 $3,500,000 9 $3,600,000 40.0% $2,400,000 $6,000,000 10 $3,000,000 50.0% $3,000,000 $6,000,000 11 $2,762,620 50.0% $2,762,620 $5,525,240 12 $3,000,000 50.0% $3,000,000 $6,000,000 MEANS $3,164,273 36.6% $1,815,852 $4,980,125 MEDIANS $3,000,000 37.5% $1,680,000 $5,756,380 ST. DEV. $1,224,743 0.114 $839,744 $1,478,538 TABLE 7 MCKEY V. TORINO PIZZA CO., INC. VARIANT ("SUNSHINE") SAMPLE DATA JURIES PRODUCING AN AWARD FOR PLAINTIFF 1 2 3 4 5 6 No. PANEL DATE VERS. HU TORINO % 1 B103 10/5 V3 60.0% 2 I206 10/31 V3 55.0%

3 C107 9/6 V3 40.0% 4 I208 10/31 V3 50.0% 5 B108 10/5 V3 49.0% 6 C110 9/6 V3 44.0% 7 G103 10/5 V3 75.0% 8 C111 9/6 V3 40.0% 9 E204 9/11 V3 40.0% 10 E206 9/11 V3 50.0% 11 F201 9/20 V3 52.0% 12 U103 1/23 V3 15.0% 13 E205 9/11 V3 10.0% 14 E203 9/11 V3 51.0% 15 U105 1/23 V3 8.0% 16 G105 10/5 V3 20.0% 17 F204 9/20 V3 48.0% MEANS 41.6% MEDIANS 48.0% ST. DEV. 0.183 1 $7 8 9 10 No. TOR.-DAMAGES STATE ST-DAMAGES TOR+ST $ % $ % 1 6,000,000 10.0% 1,000,000 70.0% 2 3,850,000 10.0% 700,000 65.0% 3 2,400,000 20.0% 1,200,000 60.0% 4 2,977,486 10.0% 595,497 60.0% 5 2,524,625 10.0% 515,230 59.0% 6 2,640,000 25.0% 1,500,000 69.0% 7 5,519,861 0.0% 0 75.0% 8 1,400,000 10.0% 350,000 50.0% 9 1,700,000 20.0% 850,000 60.0% 10 2,262,620 20.0% 905,048 70.0% 11 1,458,937 10.0% 280,565 62.0% 12 300,000 36.0% 720,000 51.0% 13 172,010 41.0% 705,241 51.0% 14 892,316 1.0% 17,496 52.0% 15 64,000 43.0% 344,000 51.0% 16 1,176,471 31.0% 182,353 51.0% 17 105,648 2.0% 4,402 50.0% MEANS $2,022,656 17.6% $580,578 59.2% MEDIANS $1,700,000 10.0% $595,497 60.0% ST. DEV. $1,816,793 0.136 $431,594 0.083 1 11 12 13 14 No. TOR.-DAMAGES STATE ST-DAMAGES TOR+ST $ % $ % 1 $7,000,000 30.0% $3,000,000 $10,000,000 2 $4,550,000 35.0% $2,450,000 $7,000,000 3 $3,600,000 40.0% $2,400,000 $6,000,000

4 $3,572,983 40.0% $2,381,989 $5,954,972 5 $3,039,854 41.0% $2,112,441 $5,152,295 6 $4,140,000 31.0% $1,860,000 $6,000,000 7 $5,519,861 25.0% $1,839,954 $7,359,815 8 $1,750,000 50.0% $1,750,000 $3,500,000 9 $2,550,000 40.0% $1,700,000 $4,250,000 10 $3,167,668 30.0% $1,357,572 $4,525,240 11 $1,739,502 38.0% $1,066,146 $2,805,648 12 $1,020,000 49.0% $980,000 $2,000,000 13 $877,251 49.0% $842,849 $1,720,100 14 $909,813 48.0% $839,827 $1,749,640 15 $408,000 49.0% $392,000 $800,000 16 $300,000 49.0% $288,235 $588,235 17 $110,050 50.0% $110,050 $220,100 MEANS $2,606,234 40.8% $1,492,415 $4,095,650 MEDIANS $2,550,000 40.0% $1,700,000 $4,250,000 ST. DEV. $1,978,053 0.083 $845,457 $2,758,252

~~~~~~~~

By (*) JORDAN H. LEIBMAN, (**) ROBERT B. BENNETT JR. and (***) RICHARD FETTER

(*) Professor of Business Law, Indiana Universit

(**) Associate Professor of Business Law, Butler Universit

(***) Associate Professor of Marketing, Butler University